Environmental footprint assessment of building ...

Building and Environment 104 (2016) 162e171

Contents lists available at ScienceDirect

Building and Environment journal homepage: www.elsevier.com/locate/buildenv

Environmental footprint assessment of building structures: A comparative study € rn Frostell a Rajib Sinha a, *, Maria Lennartsson b, Bjo a b

Industrial Ecology, KTH Royal Institute of Technology, Stockholm, Sweden City of Stockholm, Stockholm, Sweden

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 March 2016 Received in revised form 6 May 2016 Accepted 8 May 2016 Available online 10 May 2016

Following the failure to implement a rather sophisticated Excel-based environmental assessment tool, environmental load profile (ELP) in the Swedish construction industry, the City of Stockholm further developed a simplified version focusing on materials to make the tool user friendly and simple, aiming at educating stakeholders in the design phase of building construction. This study evaluated whether this simplified ELP of building structures (ELP-s) can be used directly or modified for use as a simple standard model for calculating the environmental footprint of building structures. ELP-s was compared with the two leading commercial LCA softwares, GaBi and SimaPro, based on two reference buildings: (i) a concrete and (ii) a wooden building, in order to examine the importance of material selection and the simplification of the tool. The results showed that the estimated energy footprint obtained using ELP-s was close in value to that produced by GaBi and SimaPro, but that carbon footprint was much lower with ELP-s. This great deviation in carbon footprint can be explained by the lower GHG emissions intensity per unit energy in Sweden compared with the world average or European average, the major data sources on which estimations in GaBi and SimaPro are based. These results indicate the importance of exercising care when applying commercial software tools to a specific situation in a specific country. They also indicate that the model should fit the purpose. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Environmental assessment tools Environmental footprints Life cycle assessment Buildings Environmental load profile

1. Introduction Life cycle assessment (LCA) is a widely accepted analytical tool that provides a holistic environmental perspective on a product by assessing impacts and resources used throughout its life cycle [48,26]. Following a similar life cycle inventory (LCI) based approach as LCA, increasing use of footprinting, e.g., carbon footprint, has been observed [19,20,52]. Recently, the EU [18] proposed the environmental footprints of products, with the aim of harmonizing the LCA methodology. In the present study, environmental footprints were chosen in order to be semantically consistent with the EU proposal. Environmental footprints of the built environment is an important issue for municipalities, developers and construction

* Corresponding author. Division of Industrial Ecology, Department of Sustainable Development, Environmental Science and Engineering (SEED), KTH Royal Institute of Technology, Teknikringen 34, 10044 Stockholm, Sweden. E-mail address: [email protected] (R. Sinha). URL: http://www.kth.se http://dx.doi.org/10.1016/j.buildenv.2016.05.012 0360-1323/© 2016 Elsevier Ltd. All rights reserved.

companies due to growing environmental awareness. The building sector, in Europe and globally, accounts for around 40% of the total energy use, more than one third of green house gas (GHG) emissions, 30% of raw materials use, 25% of water use, 12% of land use, and 25% of solid waste generation [5,50,51]. Therefore, the building sector needs to devote great attention to reducing its environmental footprints. Life cycle thinking [22,30] in the building sector is increasing [7,11,17,31e33,35,44]. The life cycle of buildings can be divided into three important parts: construction, use/operational phase, and demolition. Many studies have found that the operational phase accounts for most of the environmental impact during a building’s life cycle. For example, energy use in the operational phase of buildings is approximately 85% of the total [1,2,7]. Thormark [47] argues that the operational energy use in a building could be reduced by improving insulation and technical solutions and that with energy efficient solutions, the embedded energy in the building could account for 40% of the total life cycle energy. Since that study was published the construction sector has improved the energy efficiency of housing and, consequently, has

R. Sinha et al. / Building and Environment 104 (2016) 162e171

significantly increased the share of embedded energy. Thormark [47] also estimates that substitution of materials used in buildings could decrease embedded energy use by approximately 17%. In addition, numerous studies report a significant effect on the environment of material choice during the construction phase [9,39,43,53]. Building materials e the materials used in a building body e are thus an increasingly important part of the overall environmental footprints of buildings. In discussions on environmental footprints, a life cycle approach is increasingly used, i.e., taking into account all emissions to the environment, no matter where they occur. Historically, life cycle approaches to assessment of the built environment started in early 1990s [6], at that time generally in the form of different types of checklists and criteria analysis. More recently, a number of LCA based softwares/tools have been developed especially to assess the built environment, e.g., Athena [49], Building environment assessment tool (BEAT) [42], EcoEffect [3], Envest 2 [13], Environmental Load Profile (ELP) [21], Eco-Quantum [27], and Sustainable Building [15,16]. Several other softwares (e.g., SimaPro, GaBi) are available for calculating of environmental footprints/impacts in a life cycle perspective. A problem with these is that they require the purchase of costly licenses and involve much work to perform a life cycle assessment. € stad, the City of StockFor use in its eco-village Hammarby Sjo holm previously developed the so-called Environmental Load € stad [6,21], an Excel-based analytProfile (ELP) for Hammarby Sjo ical model that provides the environmental footprints of a building in a life cycle perspective. The tool has been used in many pilot cases [6,7,21], but the City of Stockholm has been facing difficulties in implementing the tool in practice [6,29,38]. This is because the tool is perceived as complex and, in some cases, too detailed for users’ actual needs. Thus, a previous attempt to simplify the tool (ELP-light) was made based on important contributing aspects to buildings that could capture 92e100% of the environmental footprints [6]. ELP-light has not yet been successfully implemented [29,38]. The main barriers to implementation of ELP and of LCA-based tools are complexity, reliability of the tool, time, and costs [6,24,32]. Based on the benefits of material selection discussed above, a second attempt to further simplify the tool was made, with the aim of educating designers (e.g., architects) about environmental footprints of material selection and related transport. The present study focuses on this simplified ELP of building structures model, which could possibly also be used as an approximate model for calculating the environmental footprints of the construction of building structures. For simplicity, the simplified ELP of building structures (Fig. 1) is referred to as ELP-s hereafter in this study. The aim of this study was to investigate and evaluate whether ELP-s of building structures can be used directly or refined for use

163

as a simple standard model for calculating the environmental footprints of building structures. Specific objectives of the study were to: Compare ELP-s with SimaPro and GaBi, by performing the same calculations for reference buildings in SimaPro and GaBi to compare the database and the algorithms Compare two reference buildings e a concrete frame building and a wooden frame building e to determine the importance of material selection Evaluate the results and determine whether ELP-s can be used for standard calculations of the environmental footprints of building structures, possibly with minor changes.

€ stad and ELP 2. Hammarby Sjo €stad is a newly developed residential city district Hammarby Sjo in Stockholm, Sweden, containing approximately 11,000 apartments and housing for approximately 35,000 people [28]. The development plan for the city district was initiated in the early 1990s because of the increasing housing demand in Stockholm [6,28]. During mid 1990s, some leading politicians in Stockholm were strongly interested in hosting Olympic games in 2004 [28] and €stad as the site for the Olympic village. suggested Hammarby Sjo The Olympic committee specified priority for the environment in its call and thus inspired by the call and the Brundtland Report [8], € stad the politicians of Stockholm decided to develop Hammarby Sjo as a forerunner for an ecologically sustainable city district [6,28,37]. Although Stockholm was not successful with its Olympic applica€ stad tion, it was decided to continue development of Hammarby Sjo with its environmental program [37]. The environmental goal of city development was to reduce the environmental load by 50% compared with the average value in the reference situation of the city of Stockholm in 1990 [6]. To make the city district twice as good, the ELP tool was developed to follow up/ €stad [6,7,21]. monitor the environmental targets for Hammarby Sjo Fig. 1 illustrates the different levels (i.e., individual, household, building, estate, and district) and activities (i.e., construction, operation, and demolition) included in the ELP tool. ELP for the built environment (ELP-full) includes all the levels and activities. The simplified version of ELP (ELP-light) is based on the subactivities contributing most to the environmental load and limits the tool to the building level (c.f., lighter background in Fig. 1). The tool was further simplified to ELP-s with the aim of educating stakeholders associated with the design and construction of buildings e property developers, consultant and architects, construction companies, constructors, and engineers. The

Fig. 1. The different activities and levels included in the ELP [6,7,21]. The dark background represents the full ELP, the lighter background the simplified version for building level, ELP-light. The white background represents the simplified ELP of building structures (ELP-s), which is tested in this study.

164


simplification was limited to account for only material inflows in construction activities at building level, shown as ELP-s in Fig. 1. This simplified version also allows a track record to be kept and construction activities to be monitored. Fig. 2 presents the structure of ELP-s in Microsoft Excel, which consists of three pages. First, it takes input for construction materials and transport from the final production site to the building construction site. In the second page, the results are shown, presenting the material flows with the environmental footprints. The final page deals with data on energy use and emissions intensities associated with construction materials and transport. The environmental footprint was estimated by adding up the footprints of the production of construction materials in a life cycle perspective and of transportation of the materials from the final production sites to the construction site. The environmental footprint for different materials was calculated using the equation:

EFi ¼

X X Mj *IP;j þ Tj *IT;j

(1)

where, EF represents environmental footprints; EFi represents the specific environmental footprint, e.g., energy footprint, carbon footprint; Mj is the amount of construction material j in mass or volume; IP,j represents the intensity of environmental footprint i to produce material j in a life cycle perspective; j represents the construction materials, e.g., concrete; Tj is the distance between the final production of the construction material j and the construction site; and IT,j represents the intensity of environmental footprint i of the transportation of material j to the construction site. ELP-s is very simple and easy to use compared with the marketleading commercially available LCA-based tools, e.g., SimaPro and GaBi. Since ELP-s is developed in Excel, it allows the user to investigate the interrelations between inputs, the database, and the results. Therefore, ELP-s is transparent and clear in calculations. Furthermore, it does not require a new complex software environment to be learned, because it can be assumed that people dealing with quantitative analysis have good knowledge of Excel, and it does not require a high level of knowledge on LCA. Building a model in ELP-s is easy and less time consuming. A student tasked with learning ELP-s, SimaPro, and GaBi, in order to test the complexity of ELP-s, required less than 1 h to learn ELP-s, approximately 30 h for SimaPro, and approximately 28 h for GaBi. This level of simplicity of ELP-s could reduce the cost and time barrier, as well as the knowledge required to learn complex software, e.g., SimaPro, GaBi. The next task was to determine the data credibility and the reliability of ELP-s, as good results in this regard could lead to acceptance of the tool by all kinds of stakeholders involved in building construction. It could also create a common platform for discussion and facilitate broader life cycle thinking among different stakeholders, e.g., architects, contractors, and engineers. 3. Methods The credibility and consistency of the ELP-s tool were examined

to determine whether they can facilitate implementation of the tool in practice and, ultimately, transition towards more environmentally benign building construction. This was done in a comparative study involving ELP-s, GaBi, and SimaPro. To examine the significance of the simplification approach in ELP-s (i.e., material selection), two types of buildings were selected: one concrete building and one wooden frame building. 3.1. LCA tools A professional market for tools/software for conducting LCAbased studies has grown up in the past 20 years [10]. Examples of fully featured LCA softwares are GaBi, SimaPro, OpenLCA, and Umberto. Nowadays, many softwares enter and vanish from the market every year [12]. For several years, SimaPro and GaBi have been dominating the market and perhaps have the greatest market share globally [10]. In the past, LCA-based studies were generally performed in spreadsheet models and this is still a powerful and potential tool for performing an LCA. Thus, GaBi and SimaPro were selected for comparison with ELP-s. 3.2. ELP-s ELP is an Excel-based tool based on the Swedish construction context. The full ELP can easily be modified by introducing external data, e.g., adding new inventories of building materials and modifying the existing database, while ELP-s is limited to building construction. It is free to use. 3.3. GaBi GaBi is a Windows-based LCA tool and developed by PE International [23]. It requires costly licenses, but supports almost all types of LCA studies. However, it requires a high degree of knowledge of LCA. The tool allows users to build models manually by drawing with very intuitive material and energy flows in the LCA. It also offers a drag and drop option to build models. Although the overall design of GaBi makes the tool intuitive, it generally takes more time to complete the model (compared with ELP-s and SimaPro). The tool calculates LCA systems in a sequential algorithm, i.e., one after another according to the input. This approach provides the benefit of running a partial model (i.e., does not need to complete the full model) to get feedback. GaBi hosts several databases, e.g., GaBi professional and EcoInvent. 3.4. SimaPro SimaPro is also a Windows-based LCA tool and developed by [45]. Like GaBi, SimaPro requires costly licenses and a high PRe degree of knowledge of LCA to learn. Building an LCA model in SimaPro is usually faster than in GaBi, but slower than in ELP-s. In addition, the tool uses matrix inversion calculation, which is a highly efficient algorithm allowing thousands of processes in one calculation [10]. Furthermore, SimaPro allows customers to install

Fig. 2. Presentation of the simplified ELP of building structures (ELP-s) in Microsoft Excel. The rectangles represent the pages in Excel and the numbers (1e3) indicate the page order.


it on an internet/intranet server, a local server, or even on a desktop/laptop computer [10,45]. SimaPro hosts a number of databases, e.g., EcoInvent, ELCD, LCAfood, ETH-ESU, US LCI, and IVAM.

Table 2 Materials used for construction of the concrete and wooden buildings. Concrete building

3.5. Environmental footprints It has recently been recognized that there is an urgent need to collect information for accounting and reporting practices in terms of environmental footprint to make LCA studies more operational [22]. Several methodological approaches have been discussed, e.g., ecological footprint [52], carbon footprint [19]. However, the scientific stringency of these approaches is arguable and sometimes inadequate. Recently, the European Commission published the Product Environmental Footprint to harmonize methodology for calculating the environmental footprints of products under the initiative Single Market for Green Products [18,20]. In order to be consistent with the EU semantic, environmental footprints were used to estimate indicators in this study. However, in contrast to the product environmental footprint, we focused on individual footprints, e.g., carbon footprint, energy footprint, water footprint, and others. In addition, we followed the methodological principles for life cycle inventories (LCI) of material and energy flows and stocks [26]. The environmental footprints of a concrete frame and a wooden frame building were estimated in a life cycle perspective using ELPs, GaBi 6, and SimaPro 8 and the results compared. A cradle-to-gate approach was used in the calculations, i.e., material production in a life cycle perspective and transport to the construction site gate. The modeling in SimaPro and GaBi was performed based on the identical processes/inventories in ELP-s or the processes that were closest to the Swedish case in the available databases. The EcoInvent 2.2 and GaBi 6 professional databases were used in GaBi 6 and the EcoInvent 2.2, ETH-ESU 96, and US LCI 2013 databases were used in SimaPro 8. The study focused only on energy footprint (kWh) and carbon footprint (kg CO2eqv) to make comparisons, although ELP-s can estimate other environmental footprints: eutrophication (g O2eqv), acidification (mol Hþeqv), photochemical ozone creation (g C2 H2eqv), and radioactive waste (cm3).

165

Concrete Concrete precast Sweden Steel virgin material Steel, Recycled Stainless steel Aluminum, virgin materials Aluminum, recycled Planed timber Plywood Particleboarda Glulamb Plasterboardc Stone woold PP/PEe PS Polystyrene Glass 100% raw material IGUf Total

Wooden building

kg

%

kg

%

3,100,000 1,100,000 6800 79,000 2600 2500 2500 44,000 3100 4000 53 000 20,000 28,000 8900 32,000 19,000 20,000 4,600,000

69% 24% 0% 2% 0% 0% 0% 1% 0% 0% 1% 0% 1% 0% 1% 0% 0% 100%

0 0 700 0 0 0 0 75,000 29,000 0 379 000 173,000 43,000 880 0 0 15,000 720,000

0% 0% 0% 0% 0% 0% 0% 10% 4% 0% 53% 24% 6% 0% 0% 0% 2% 100%

a Engineered wood products from sawmill dusts, sawmill savings, and wood chips. b Glued laminated timber. c Known as drywall produced from thick sheets of paper and gypsum plaster. d Fiber materials from rock. e Polypropylene/polyethylene. f Insulated glass units.

for entries that were reviewed by energy experts at the City of Stockholm. For estimating the environmental footprints, we only considered the main structures of the buildings (i.e., not including the detailed materials used in the buildings, e.g., materials for interior design), as shown in Table 2. We also excluded the foundations of the two buildings. The developers concerned provided material and transport data related to the building constructions.

4. Results 3.6. Case selection The reference buildings used (a wooden frame and a concrete frame building) were two of the best entries in a competition in which the City of Stockholm invited different developers to submit plans for energy-efficient buildings. The two buildings have quite similar total floor areas, as shown in Table 1, as the competition set the same requirements on all entries regarding performance and place (the same plot of land). The functionality of the two buildings was also the same, as they had to be built to the same standard regarding fire performance, noise barriers, installations, etc. as regulated by the Swedish building code and the Stockholm Royal Seaport special requirements, which were part of the conditions for the competition. The targeted energy performance was regulated for the competition, the baseline being that the building would generate more energy than it consumed, calculated over one year. This requirement was verified by energy performance calculations

Table 1 Building specifications for the concrete and wooden building.

Total floor area (BTA) Facade area Window area Roof area

Concrete building (m2)

Wooden building (m2)

4500 2599 651 900

4925 3520 804 920

A comparison of energy and carbon footprints between ELP-s, GaBi, and SimaPro for the concrete and wooden building is presented in Table 3. In ELP-s, the calculation for unit floor area showed around 43% lower energy footprint and 75% lower carbon footprint for the wooden building than the concrete building. Compared with the ELP-s values, energy footprint in GaBi and SimaPro differed by 10% and 7%, respectively, for the concrete building and 109% and 7% respectively, for the wooden building. For carbon footprint in GaBi and SimaPro compared with ELP-s, the values for the concrete building were 57% and 41% higher, respectively, and those for the wooden building were 92% and 68% higher, respectively. For both buildings, SimaPro showed an energy footprint close to ELP-s (7% higher), where as the carbon footprint was around 50% higher than with ELP-s. These higher carbon footprints in SimaPro can be explained by the emissions intensity per energy use in Sweden compared with the world. ELP-s was based on materials produced in Sweden, whereas most of the data used in the SimaPro model for both buildings were European average value. In the EcoInvent database 2.2 (used in both GaBi and SimaPro), GHG emissions per unit energy (Swedish electricity grid mix) are around 80% lower than the European average.1

1 The average electricity grid mix of Germany, Denmark, Switzerland, Estonia, Finland, France, Italy, and Austria.

166


Table 3 Comparison of energy and carbon footprints obtained using ELP-s, GaBi, and SimaPro for the concrete and wooden buildings. Environmental footprints

Energy (MWh)

Carbon kg CO2eqv

Energy (%) (varies from ELP-s) Carbon (%) (varies from ELP-s) a

Concrete building

ELP-s GaBi SimaPro ELP-s GaBi SimaPro GaBi SimaPro GaBi SimaPro

Wooden building

Total

/m2 BTAa

Total

/m2 BTAa

3700 4100 4000 720,000 1,100,000 1,000,000 10% 7% 57% 41%

0.820 0.91 0.89 160 240 220 10% 7% 57% 41%

2300 4800 2500 200,000 380,000 330,000 109% 7% 92% 68%

0.47 1 0.50 40 76 67 109% 7% 92% 68%

BTA ¼ total floor area.

For both buildings, the GaBi results showed higher values for energy and carbon footprints than the SimaPro results. In GaBi, conversion factors were used to convert the wooden materials from volume to weight. The conversion factors were taken from ELP-s, since we could not find the relationship between volume and weight in GaBi. The conversion might have caused the higher values in the GaBi results compared with the SimaPro results. For example, glulam in GaBi (see the figures below) showed very high energy and carbon footprints compared with ELP-s, while SimaPro showed closer energy footprint to ELP-s. 4.1. Concrete vs wooden building The most important materials on a weight basis were concrete (93% of total weight) for the concrete building and glulam (53%), plasterboard (24%), and planed timber (10%) for the wooden building. The wooden building required around six times less materials in terms of weight than the concrete building (Table 2). If the foundations had been considered in the material inventories for analysis of the two buildings, the concrete building would have needed around six times stronger foundations than the wooden building. Consequently, the weight of the concrete building was much greater than that of the wooden building. 4.2. Concrete building A comparison between ELP, GaBi, and SimaPro of the energy footprint of the materials used in construction of the concrete building is shown in Fig. 3a. As can be seen from the diagram, concrete, steel and PS polystyrene had a larger energy footprint than other materials needed for construction of the concrete building. The energy footprint of the three most contributing materials in ELP-s, GaBi, and SimaPro was 74%, 62%, and 66%, respectively, of total energy footprint of the building. The energy footprint from transport was around 3% in ELP-s and GaBi and 6% in SimaPro. Among wood products (7% of total in ELP-s and SimaPro), glulam2 and timber were the main contributors to the energy footprint (5% in ELP and GaBi; see Fig. 3a). A comparison between ELP, GaBi, and SimaPro for carbon footprint of the materials used in construction of the concrete building is shown in Fig. 3b. Concrete, steel, and PS polystyrene had a larger carbon footprint than other materials needed for construction of the concrete building. The carbon footprint of the three most

2 Glulam in GaBi was not correctly converted (in our opinion) from volume to weight; thus it resulted in very high energy and carbon footprints compared with ELP-s and SimaPro. For this reason, we did not include GaBi in comparisons for wooden products.

contributing materials in ELP-s, GaBi, and SimaPro was 82%, 84%, and 80%, respectively, of the total carbon footprint of the building. Carbon footprint from transport was around 4% in ELP-s and GaBi and 5% in SimaPro. Wood products made a very negligible contribution to carbon footprint (around 1% of the total). 4.3. Wooden building A comparison between ELP, GaBi, and SimaPro for energy footprint of the materials used in the construction of the wooden building is presented in Fig. 4a. Wood products used more energy than other materials needed for construction of the wooden building. The energy footprint of the wood products in ELP-s, GaBi, and SimaPro was 67%, 84%, and 54%, respectively, of the total energy footprint of the building. Among wood products, glulam contributed around 43% of the total energy footprint in both ELP-s and GaBi. Energy footprint for transport was around 7% in ELP-s and GaBi and 19% in SimaPro. Plasterboard and stone wool gave 20% energy footprint in ELP-s and SimaPro and 8% in GaBi (very low because of the high contribution from glulam in GaBi). A comparison between ELP-s, GaBi, and SimaPro for carbon footprint of the materials used in construction of the wooden building is made in Fig. 4b. Gulam, plasterboard, stone wool, and transport had a significantly larger carbon footprint than other materials used in the construction of the wooden building. Glulam contributed around 15% and 28% in ELP-s and SimaPro, respectively. Carbon footprint from truck transport of the materials was around 27%, 19%, and 30% in ELP-s, GaBi, and SimaPro, respectively. 4.4. Material used in the buildings Considering results for the concrete and the wooden building, concrete, steel, plywood, glulam, plasterboard, stone wool, PS polystyrene, and truck transport contributed significantly to the energy and carbon footprint (around 90% of the total). Fig. 5 compares the energy and carbon footprints per unit weight (i.e., per kg) of materials used in the buildings according to ELP-s, GaBi, and SimaPro. In this comparison, concrete (cf., Fig. 3), plasterboard, stone wool, and PS polystyrene showed less variation in the results from GaBi and SimaPro compared with ELP-s. Steel showed very high variation in the results from GaBi and SimaPro compared with ELP-s (Fig. 5). ELP-s also showed some contradictory data for steel in the database, for example, the energy and carbon footprints of stainless steel cannot be lower than those of virgin steel material (cf. ELP-s for steel virgin material and stainless steel in Fig. 5). Furthermore, recycled steel in ELP-s showed similar energy footprint and GHG emissions to virgin steel. In GaBi, recycled steel showed even higher energy and carbon footprints than virgin steel. However, the literature suggests that


167

Fig. 3. a) Energy footprint and b) carbon footprint per unit floor area for the concrete building according to ELP-s, GaBi, and SimaPro.

recycled steel generally takes much less energy than virgin steel (e.g., [40]). Data in GaBi for steel and recycled steel were obtained from EcoInvent (version 2.2) global and European average respectively, and the data origin and the assumption behind the LCA model could explain the discrepancy. From a life cycle perspective, virgin aluminum showed higher energy and carbon footprint per kg than other building materials (Fig. 5), though it made a low contribution to the total energy footprint of the concrete building (Fig. 3a). In addition, aluminum showed a high variation in both energy and carbon footprint between ELP, GaBi, and SimaPro. Based on the above findings, we

concluded that metals in ELP-s show higher variations than other building materials. The estimated energy footprint for glulam was very similar in ELP-s and SimaPro. However, it showed high variation in GaBi compared with ELP-s and SimaPro. The conversion error mentioned earlier might be the reason for the large variation. We also compared energy and carbon footprint per ton-km of transport between ELP-s, GaBi, and SimaPro, as shown in Fig. 6. Energy and carbon footprint for each mode of transportation showed large variation in all three models. Comparing the different transport modes available in ELP-s, train transport showed the

168


Fig. 4. a) Energy footprint and b) carbon footprint per unit floor area for the wooden building according to ELP-s, GaBi, and SimaPro.

largest variation between the software types. Train transport per ton-kilometer in ELP-s showed very low energy and carbon footprint compared with in GaBi and SimaPro. It is possible that ELP-s did not allocate the contribution from the life cycle inventory of manufacturing trains. Thus, it is strongly recommended that train transport in ELP-s be revised. For cargo ship transport, GaBi showed very low energy and carbon footprint compared with ELP-s and SimaPro.

release humidity, absorb unwanted noise, has good thermal conductivity, is warm to the touch, and so on [36]. The present study did not include the operation and demolition stages of the buildings. The GHG balance of a wooden building would be affected strongly by how the waste is handled after demolition, and whether the forest would have managed to re-grow the wood [4,34]. To sum up, use of wood-based materials would be a better strategy for lowering the environmental footprints of buildings with similar functionality.

5. Discussion 5.2. ELP-s, GaBi and SimaPro 5.1. Concrete vs wood As already discussed, the wooden building (per unit floor area) had around 40% lower energy footprint and around 70% lower €jesson and Guscarbon footprint than the concrete building. Bo tavsson [4] found that wooden frame buildings required about 60e80% lower primary energy input (in production of building materials) than concrete frame buildings. A possible explanation for the low energy footprint of the wooden building in our study compared with the literature could be the omission of the foundations and detailed materials (e.g., material for interior design) required for the construction. Pajchrowski et al. [36] compared four types of buildings with similar functionality, and concluded that wood and wood-based material had environmental benefits both from cradle-to-gate and gate-to-grave. Other studies have also found that wood-based materials reduce over all carbon footprint and are sometimes cost-effective [4,34,41]. In addition, wood has been attracting attention for modern constructions as a light-weight, mechanically strong, renewable material that can adsorb and

The results for energy footprint in a life cycle perspective for both buildings showed satisfactorily close outcomes in GaBi, SimaPro, and ELP-s. However, individual materials showed great variation in different tools. In order to use ELP-s for standard calculations of building environmental footprint, the data it contains for a number of materials need to be revised. For example, data for stainless steel in ELP-s need to be modified since they currently result in a lower environmental footprint than virgin steel, contradicting the literature. We investigated the environmental product declaration (EPD) for the energy and carbon footprint of stainless steel. According to an EPD [14] for a Nordic company (Outokumpu Oyj), the energy footprint and carbon footprint of stainless steel are 11.41 kWh/kg and 2.75 kg CO2eqv/kg respectively. Comparison of the EPD with ELP-s, GaBi, and SimaPro results indicated that the stainless steel data in ELP-s require revision (cf. Fig. 5). The EDP data showed a good agreement with the Swedish case, as well as with the virgin steel in ELP-s. We replaced the data for stainless steel in ELP-s with


169

Fig. 5. a) Energy footprint and b) carbon footprint per kg of building materials according to ELP-s, GaBi, and SimaPro.

the EPD value and made an additional test to see how this influenced the total footprint of the concrete building. We found that it contributed only a 1% increase in the total energy and carbon footprint due to the very small share of stainless steel in the concrete building. Since the glulam in GaBi showed inadequate results, we made an extra test by correcting the glulam conversion in GaBi using the SimaPro data for glulam to see the influence on the total footprint of the building structures. The concrete building showed 9% error in energy footprint and 1% error in carbon footprint. Since glulam comprised a larger share of the wooden building, the correction greatly affected the results, showing 91% error in energy footprint and 20% error in carbon footprint. The results thus indicate the importance of (i) being careful with conversions and calculations outside the software environment and (ii) software developers and providers giving more comprehensive information on the

processes in their LCI databases. We also recommend that tool developers provide users with (i) a flexible environment of unit conversion available in practice and (ii) better possibilities to access the comprehensive information about the processes in their databases. It was very difficult to find case-specific data in the databases used in GaBi and SimaPro. Sometimes data for the same material and from similar origins and similar technology were used, but the models produced different results. Herrmann and Moltesen [25] investigated whether it matters which LCA tool is chosen through a comparative assessment of SimaPro and GaBi for identical processes taken from identical databases and versions. They found that the results were identical in many cases but also that the differences were sometimes so great that the conclusions could be doubted. In an extreme case, they found a 109-fold difference between the model results. Speck et al. [46] also discovered that using

170


6. Conclusions

Fig. 6. a) Energy footprint and b) carbon footprint per ton-km transport of the building materials according to ELP-s, GaBi, and SimaPro.

common methods with closely matched inputs could provide differing results in different softwares. ELP-full and ELP-s are based on Swedish data but in the available databases for GaBi and SimaPro, we rarely found data related to the Swedish case. Comparisons of GaBi and SimaPro showed that there was no exactly identical environmental footprint from the materials in the buildings (cf. Fig. 5), although a few materials (concrete, glass raw material, and IGU) showed nearly identical results for GaBi and SimaPro. In other cases, the comparison showed differences in environmental footprint between the softwares that were sometimes comparatively large (cf. plywood in Fig. 5). We also found that specific process in two different databases did not produce identical environmental footprints. For example, a specific process for plywood was taken from two databases, (i) GaBi professional in GaBi and (ii) EcoInvent in SimaPro, but the energy and carbon footprints of the plywood in GaBi and Simapro were quite different (cf. Fig. 5). Based on the above discussion, ELP-s may allow designers to make standard calculations of the environmental footprint of building structures and can familiarize them with (or even educate them about) environmental footprints of their material selections in a very easy way compared with SimaPro and GaBi or other related software.

5.3. Limitations and future works ELP-s is based on old data produced around a decade ago and detailed new inventory modeling of each product, or at least for a few significantly contributing materials, e.g., concrete, glulam, is required for the Swedish case. The ELP database is suitable for the materials produced in Sweden. For materials produced outside Sweden, new inventories would be needed, using country- and technology-specific LCIs. ELP-s can be improved by including more alternative materials, e.g., based on different technologies and different countries, to allow the user to compare and chose a suitable option from alternatives. Transport options and data also need to be revised with detailed LCIs of transport options and inventories for multiple alternatives, e.g., different types of trucks.

This study examined whether ELP of building structures (ELP-s) can be used as a simple standard model for calculating the environmental footprint of building structures. One concrete and one wooden building were selected to compare ELP-s with two other commercially available market-leading softwares, GaBi and SimaPro. Comparisons of the two buildings and their constituent materials showed that the wooden building had a lower environmental footprint than the concrete building. The estimated energy footprint according to ELP-s was close to the values from GaBi and SimaPro (around 10% and 7% difference for GaBi and SimaPro, respectively). However, carbon footprint was around 50% higher with GaBi and SimaPro compared with ELP-s. This proved to be because ELP-s is based on the materials being produced in Sweden, while most of the data used in SimaPro for both buildings were European average values. In the EcoInvent database, GHG emissions per unit energy used (Swedish electricity grid mix) are much lower than the European average. Therefore, the higher carbon footprint in SimaPro and GaBi compared with ELP-s was because of the lower emissions intensity per energy use in Sweden compared with the world/Europe. However, no strong relations and patterns emerged in detailed analysis of each material in the buildings using the three models. Thus, it is important to exercise care in the use of commercial software tools when applying them to a specific situation in a specific country, e.g., considering the carbon intensity of its energy system. There were important differences in results between GaBi and SimaPro, indicating potential for improving the software to assure consistency. Further, it is recommended that LCA-based studies focus on data specific to the case and use a model suitable for the specific purpose. For instance, ELP-s is based on Swedish data and the city of Stockholm intends to use it to educate/engage designers to include environmental footprints in their design phase of buildings. This study concluded that after minor modifications, ELP-s can be used for standard calculations of the environmental footprint of building structures and can create a common platform for discussions among stakeholders about building construction. Acknowledgments We thank the City Of Stockholm (Stockholms stad) contract 2014-06-25 for financial support for the project. Rajib Sinha and €rn Frostell also gratefully acknowledges Industrial Ecology, KTH Bjo Royal Institute of Technology, Stockholm, Sweden, for financial support to transform the project results to a scientific publication. References [1] K. Adalberth, Energy use during the life cycle of single-unit dwellings: examples, Build. Environ. 32 (4) (1997) 321e329. [2] K. Adalberth, Energy use in four multi-family houses during their life cycle, Int. J. Low Energy Sustain. Build. 1 (1999) 1e20. [3] G. Assefa, M. Glaumann, T. Malmqvist, B. Kindembe, M. Hult, U. Myhr, O. Eriksson, Environmental assessment of building propertiesdwhere natural and social sciences meet: the case of ecoeffect, Build. Environ. 42 (3) (2007) 1458e1464. €rjesson, L. Gustavsson, Greenhouse gas balances in building construction: [4] P. Bo wood versus concrete from life-cycle and forest land-use perspectives, Energy policy 28 (9) (2000) 575e588. n, S. Scarpellini, Life cycle assessment in buildings: state[5] I.Z. Bribi an, A.A. Uso of-the-art and simplified LCA methodology as a complement for building certification, Build. Environ. 44 (12) (2009) 2510e2520. [6] K. Brick, Barriers for Implementation of the Environmental Load Profile and Other LCA-based Tools, Licentiate thesis at Industrial Ecology, KTH Royal Institute of Technology, 2008. [7] K. Brick, B. Frostell, A comparative study of two Swedish LCA-based tools for practical environmental evaluation of buildings, J. Environ. Assess. Policy Manag. 9 (03) (2007) 319e339. [8] G. Brundtland, M. Khalid, S. Agnelli, S. Al-Athel, B. Chidzero, L. Fadika, V. Hauff,


[9] [10] [11]

[12]

[13] [14]

[15]

[16]

[17]

[18]

[19] [20]

[21]

[22] [23] [24] [25]

[26] [27] [28]

[29]

[30] [31]

I. Lang, M. Shijun, M.M. de Botero, et al., Our Common Future (\’brundtland report\’), 1987. A.H. Buchanan, S.B. Levine, Wood-based building materials and atmospheric carbon emissions, Environ. Sci. Policy 2 (6) (1999) 427e437. A. Ciroth, Software for Life Cycle Assessment. Life Cycle Assessment Handbook: a Guide for Environmentally Sustainable Products, 2012, pp. 143e157. Y.H. Dong, S.T. Ng, A life cycle assessment model for evaluating the environmental impacts of building construction in Hong Kong, Build. Environ. 89 (2015) 183e191. I. Dunmade, LCA software tools and approaches, life-cycle assessment in environmental practice, in: AGM/Conference, The Learning Centre, Mount Royal College, Calgary, Alberta, 2007. Envest 2, Envest 2, homepage of Envest2, 2015. http://envest2.bre.co.uk. EPD, Hot Rolled Stainless Steel: Outokumpu Oyj, 2014. Environmental Product Declaration as per ISO 14025. (Online). Available at: http://www.outokumpu. com/SiteCollectionDocuments/EPD-OTO-20140001-IBD1-EN.pdf. M. Erlandsson, The System Sustainable Buildings’. Tech. Rep., IVL Report B1605, IVL Swedish Environmental Research Institute, Stockholm, 2005 [in Swedish]. M. Erlandsson, M. Borg, Generic LCA-methodology applicable for buildings, constructions and operation servicesdtoday practice and development needs, Build. Environ. 38 (7) (2003) 919e938. M. Erlandsson, P. Levin, Environmental assessment of rebuilding and possible performance improvements effect on a national scale, Build. Environ. 40 (11) (2005) 1459e1471. EU, Commission recommendation of 9 april 2013 on the use of common methods to measure and communicate the life cycle environmental performance of products and organisations, Off. J. Eur. Union 56 (2013). M. Finkbeiner, Carbon footprintingdopportunities and threats, Int. J. Life Cycle Assess. 14 (2) (2009) 91e94. M. Finkbeiner, Product environmental footprintdbreakthrough or breakdown for policy implementation of life cycle assessment? Int. J. Life Cycle Assess. 19 (2) (2014) 266e271. A. Forsberg, Environmental Assessment of the Urban Environmentedevelopment and First Application of the Environmental Load Profile €stad, Licentiate Thesis, Industrial Ecology, Department of for Hammarby Sjo Chemical Engineering and Technology, Royal Institute of Technology, Stockholm, 2003. B. Frostell, Life cycle thinking for improved resource Management: LCA or?, in: Handbook of Sustainable Engineering, Springer, 2013, pp. 837e857. GaBi, Software, 2015. http://www.gabi-software.com/sweden/software/. J. Giesekam, J.R. Barrett, P. Taylor, Construction sector views on low carbon building materials, Build. Res. Inf. (2015) 1e23. I.T. Herrmann, A. Moltesen, Does it matter which life cycle assessment (lca) tool you choose?ea comparative assessment of simapro and gabi, J. Clean. Prod. 86 (2015) 163e169. ISO 14044, Environmental ManagementdLife Cycle Assessmentd Requirements and Guidelines, Switzerland, Geneva, 2006. IVAM, Research and Consultancy on Sustainability, 2015. EcoQuantum, http:// www.ivam.uva.nl/en/?id¼2&L¼1. S.P. Iverot, N. Brandt, The development of a sustainable urban district in € stad, Stockholm, Sweden? Environ. Dev. Sustain. 13 (6) (2011) Hammarby Sjo 1043e1064. S.P. Iveroth, A.-L. Vernay, K.F. Mulder, N. Brandt, Implications of systems €stad, Stockholm, integration at the urban level: the case of Hammarby Sjo J. Clean. Prod. 48 (2013) 220e231. S. Kotaji, A. Schuurmans, S. Edwards, Life-cycle Assessment in Building and Construction, A state-of-the-art report, 2003, Setac, 2003. X. Li, Y. Zhu, Z. Zhang, An LCA-based environmental impact assessment model

171

for construction processes, Build. Environ. 45 (3) (2010) 766e775. [32] T. Malmqvist, M. Glaumann, S. Scarpellini, I. Zabalza, A. Aranda, E. Llera, S. Díaz, Life cycle assessment in buildings: the enslic simplified method and guidelines, Energy 36 (4) (2011) 1900e1907. [33] M. Mequignon, H.A. Haddou, F. Thellier, M. Bonhomme, Greenhouse gases and building lifetimes, Build. Environ. 68 (2013) 77e86. n, F. Hedenus, S. Karlsson, J. Holmberg, Concrete vs. wood in buil[34] J. N€ asse dingsean energy system approach, Build. Environ. 51 (2012) 361e369. [35] O. Ortiz, F. Castells, G. Sonnemann, Sustainability in the construction industry: a review of recent developments based on lca, Constr. Build. Mater. 23 (1) (2009) 28e39. [36] G. Pajchrowski, A. Noskowiak, A. Lewandowska, W. Strykowski, Wood as a building material in the light of environmental assessment of full life cycle of four buildings, Constr. Build. Mater. 52 (2014) 428e436, http://dx.doi.org/ 10.1016/j.conbuildmat.2013.11.066. €rdering av hammarby sjo €stads miljo €profiler[37] S. Pandis Iverot, N. Brandt, Utva ing: vilka erfarenheter ska tas med till nya stadsutvecklingsprojekt i stockholm?, 2009. [38] S. Pandis Iveroth, Industrial Ecology for Sustainable Urban Development-the € stad, Ph.D. thesis, Division of Industrial Ecology, KTH Case of Hammarby Sjo Royal Institute of Technology, 2014. [39] A.K. Petersen, B. Solberg, Greenhouse gas emissions, life-cycle inventory and cost-efficiency of using laminated wood instead of steel construction.: case: beams at gardermoen airport, Environ. Sci. Policy 5 (2) (2002) 169e182. [40] M. Reuter, C. Hudson, A. van Schaik, K. Heiskanen, C. Meskers, C. Hagelüken, Metal Recyling: Opportunities, Limits, Infrastructure, UNEP, International Resource Panel, Working Group on the Global Metal Flows, 2013. [41] R. Sathre, L. Gustavsson, Using wood products to mitigate climate change: external costs and structural change, Appl. Energy 86 (2) (2009) 251e257. [42] SBI, Building Environment Assessment Tool (BEAT) Developed by Danish Building Research Institute (SBI), 2015. Denmark, http://www.sbi.dk/en. [43] C. Scheuer, G.A. Keoleian, P. Reppe, Life cycle energy and environmental performance of a new university building: modeling challenges and design implications, Energy Build. 35 (10) (2003) 1049e1064. [44] A. Sharma, A. Saxena, M. Sethi, V. Shree, et al., Life cycle assessment of buildings: a review, Renew. Sustain. Energy Rev. 15 (1) (2011) 871e875. [45] SimaPro, LCA Software: about SimaPro, 2015. http://www.pre-sustainability. com/simapro. [46] R. Speck, S. Selke, R. Auras, J. Fitzsimmons, Life cycle assessment software: selection can impact results, J. Ind. Ecol. 20 (1) (2015) 18e28. [47] C. Thormark, The effect of material choice on the total energy need and recycling potential of a building, Build. Environ. 41 (8) (2006) 1019e1026. [48] A. Tillman, H. Baumann, The Hitchhikers Guide to LCA, Studentlitteratur AB, Lund, Sweden, 2004. [49] W. Trusty, J. Meil, G. Norris, Athena: a lca decision support tool for the building community, Green Build. Chall. 98 (1998) 39e46. [50] UNEP, Buildings and Climate Change: Summary for Decision-makers, United Nations Environment Programme, 2009a. [51] UNEP, Common Carbon Metric for Measuring Energy Use and Reporting Greenhouse Gas Emissions from Building Operations, United Nations Environment Programme, 2009b. [52] M. Wackernagel, W. Rees, Our Ecological Footprint: Reducing Human Impact on the Earth, New Society Publishers, 1998. No. 9. [53] P. Westlund, M. Brogren, K. Byman, B. Hylander, J. Kellner, C. Linden, € Lo € nngren, J. Nordling, L. Stro €mberg, F. Winberg, Klimatpåverkan från O. byggprocessen (in english: Environmental impact of the construction process), Tech. Rep. IVA-M 449. ISSN: 1102-8254, 2014, ISBN 978-91-7082-883-6 (En rapport från IVA och Sveriges Byggindustrier).