Int J Life Cycle Assess

DOI 10.1007/s11367-013-0686-x

BUILDING COMPONENTS AND BUILDINGS

Life cycle assessment of construction materials: the influence

of assumptions in end-of-life modelling
Gustav Sandin & Greg M. Peters & Magdalena Svanström

Received: 20 February 2013 / Accepted: 9 December 2013

# Springer-Verlag Berlin Heidelberg 2013

Abstract Results and discussion In all comparable scenarios, glulam

Purpose The nature of end-of-life (EoL) processes is highly beams have clear environmental benefits compared to steel
uncertain for constructions built today. This uncertainty is frames, except for in a scenario in which steel frames are
often neglected in life cycle assessments (LCAs) of construc- recycled and today’s average steel production is substituted,
tion materials. This paper tests how EoL assumptions influ- in which impacts are similar. The choice of methodological
ence LCA comparisons of two alternative roof construction approach (attributional, consequential or fully disregarding
elements: glue-laminated wooden beams and steel frames. EoL processes) does not seem to influence the relative perfor-
The assumptions tested include the type of technology and mance of the compared construction elements. In absolute
the use of attributional or consequential modelling terms, four factors are shown to be critical for the results:
approaches. whether EoL phases are considered at all, whether recycling or
Methods The study covers impact categories often considered incineration is assumed in the disposal of glulam beams,
in the construction industry: total and non-renewable primary whether a consequential or attributional approach is used in
energy demand, water depletion, global warming, eutrophica- modelling the disposal processes and whether today’s average
tion and photo-chemical oxidant creation. The following ele- technology or a low-impact technology is assumed for the
ments of the EoL processes are tested: energy source used in substituted technology.
demolition, fuel type used for transportation to the disposal Conclusions The results suggest that EoL assumptions can be
site, means of disposal and method for handling allocation highly important for LCA comparisons of construction mate-
problems of the EoL modelling. Two assumptions regarding rials, particularly in absolute terms. Therefore, we recommend
technology development are tested: no development from that EoL uncertainties are taken into consideration in any LCA
today’s technologies and that today’s low-impact technologies of long-lived products. For the studied product type, LCA
have become representative for the average future technolo- practitioners should particularly consider EoL assumptions
gies. For allocating environmental impacts of the waste han- regarding the means of disposal, the expected technology
dling to by-products (heat or recycled material), an attribu- development of disposal processes and any substituted tech-
tional cut-off approach is compared with a consequential nology and the choice between attributional and consequential
substitution approach. A scenario excluding all EoL processes approaches.
is also considered.
Keywords Attributional . Consequential . Construction
product . Disposal . LCA . Long-lived product .
Responsible editor: Marzia Traverso Infrastructure . Waste management
Electronic supplementary material The online version of this article
(doi:10.1007/s11367-013-0686-x) contains supplementary material,
which is available to authorized users. 1 Introduction
G. Sandin (*)
Department of Wood Technology, SP Technical Research Institute of For long-lived products manufactured today, end-of-life
Sweden, Brinellgatani 4, Box 857, 501 05 Borås, Sweden (EoL) processes such as demolition and disposal will take
e-mail: [email protected] place in a distant future. For construction materials, EoL
G. Sandin : G. M. Peters : M. Svanström
processes are often estimated to take place in 50–100 years
Division of Chemical Environmental Science, Chalmers University (Frijia et al. 2011). Due to technological change, the nature of
of Technology, Kemivägen 10, 412 96 Gothenburg, Sweden such processes is highly uncertain. This time-dependent
 Int J Life Cycle Assess

uncertainty has previously been acknowledged as a challenge in a near future and processes occurring in a distant future,
typical for life cycle assessments (LCAs) in the construction such as EoL processes, as these are inevitably subject to larger
industry (Singh et al. 2011; Verbeeck and Hens 2007). This uncertainties. Due to these uncertainties, a range of scenarios
uncertainty is nevertheless often neglected in LCAs of con- are set up to test how assumptions regarding EoL processes
structions and construction materials, and EoL practices of influence the LCA results.
today are assumed to be valid without any explicit explana-
tion, even when the aim is to support decisions 1.2 Second aim: to compare the influence of attributional
concerning contemporary constructions that are expected and consequential approaches to EoL modelling
to stand for a long time (e.g. Bribián et al. 2011; Habert et al.
2012; Lundie et al. 2004; Persson et al. 2006). There are The second aim concerns the influence of a specific assump-
exceptions; for example, Bouhaya et al. (2009) set up scenar- tion in EoL modelling: the choice between an attributional and
ios to account for different possible future means of EoL a consequential approach. A consequential approach is in-
treatment of a bridge. creasingly used in LCAs but often in an inconsistent manner.
To consider EoL uncertainties is especially important when Even LCA standards and handbooks (e.g. BSI 2011;
EoL practices may significantly influence the environmental WBCSD/WRI 2011; European Commission 2010) permit
impact. For buildings, efficient recycling at the disposal stage the use of consequential elements (e.g. substitution as a means
may save energy that corresponds to 29 % of the energy use in of avoiding allocation) in attributional LCAs. This has been
manufacturing and transportation of the construction materials criticised as leading to results with unclear meaning (Brander
(Blengini 2009). Moreover, energy savings from efficient and Wylie 2012). There is clearly a need for research on how
recycling may correspond to 15 % of the total energy use of to choose between different LCA approaches, how to apply
a building’s life cycle (Thormark 2002). Although a building’s the chosen approach consistently and in what contexts the
use phase is often said to contribute 70–90 % of its environ- choice matters. This paper aims at contributing to this re-
mental impact (Beccali et al. 2013; Cuéllar-Franca and search, by comparing attributional and consequential ap-
Azapagic 2012; Ortiz et al. 2010), the relative importance of proaches for modelling the EoL processes of construction
EoL processes is now rising due to increasingly energy- materials.
efficient buildings (Dixit et al. 2012); it has even been argued
that poorly defined functional units often lead to exaggerated
data on energy usage in the use phase (Frijia et al. 2011). The 2 Methods
environmental impact of the waste handling of construction
materials is also considered significant simply because of the LCA was used because it is a well-developed and increasingly
sheer amount of such materials existing in society (Bribián popular method (Peters 2009) for assessing the environmental
et al. 2011; Singh et al. 2011; Blengini 2009). Finally, it has impact of products. We used the LCA software GaBi 5 (PE
been shown that assumptions on EoL modelling can be of International 2013) and selected a subset of the environmental
great importance for the life cycle impact of construction impacts and resource use parameters recommended in EN
materials (Ardente et al. 2008). 15804, the standard for environmental product declarations
So there are strong reasons to improve the modelling of of construction products (SIS 2012): total and non-renewable
EoL processes in LCAs of construction materials. This can primary energy demand (PE), water depletion potential, glob-
contribute to more robust decision making in the construction al warming potential (GWP), eutrophication potential (EP)
sector, for example, in the development of new construction and photo-chemical oxidant creation potential (POCP). A
materials and policies. subset of the impact categories recommended in EN 15804
was deemed sufficient for the purpose of this paper, but for a
1.1 First aim: to test how EoL assumptions influence LCA comprehensive comparison of building elements, all recom-
comparisons of different construction materials mended impact categories should be considered along with
other relevant impact categories (as discussed in section 5).
In the present paper, we use LCA to compare the environ- The applied total PE indicator accounts for all energy
mental impact of alternative internal roof constructions for an extracted from the earth, including renewable energy and the
industrial hall: glue-laminated (glulam) wooden beams and energy content of any material and fuel inputs. This indicator
steel frames. The geographical scope is Europe, thus the reflects a concern about limited availability of energy re-
product systems are modelled for European average condi- sources in society. The non-renewable PE indicator focuses
tions. A temporal dimension is introduced in the mapping of on the exhaustibility and limited access to fossil energy
the product system, which contrasts with the traditional LCA sources (Arvidsson et al. 2012). Water depletion potential is
practice of assuming time-invariant product systems. This is the name given to the water use impact category in the ReCiPe
done by distinguishing between processes occurring today or framework (Goedkoop et al. 2012); this is a basic volumetric
Int J Life Cycle Assess

proxy for the burden on the environment and resources caused future states by assuming a single technology to be represen-
by water use. More comprehensive approaches exist (see tative for the average technology.
Kounina et al. 2013), but for accuracy, they require detailed The consequential approach is also present when substitu-
information about the location of the water being used (Sandin tion is applied to avoid allocation in the disposal stage. This
et al. 2013). GWP, EP and POCP are commonly studied means that by-products of the disposal process are assumed to
impact categories in the construction sector with established substitute a product manufactured by alternative means and
characterisation methods. For characterisation, we used the that the environmental impacts thereby avoided are credited to
CML 2001 framework (Guinée et al. 2002), with factors the system under study. In EoL scenarios with incineration as
updated in 2010; thus GWP calculations include carbon se- the assumed means of disposal, the by-product is assumed to
questered in the product and considers biogenic carbon diox- be heat, and the substituted means of heat production is
ide emissions as climate neutral (how to account for seques- assumed to be combustion of natural gas or municipal
tration and temporal storage of carbon is a topic of debate— biowaste (food and gardening waste with a water content of
see Brandão et al. 2013). 60 wt% (Jungbluth et al. 2007)). In EoL scenarios with
To properly provide understanding of methodological recycling as the assumed means of disposal, the by-product
choices regarding system boundaries and choice of data type, is the recycled material (wood and steel, respectively), and
there is a need to elaborate on where the present study is when substitution is applied, the substituted means of material
positioned on the attributional-consequential spectrum. production is assumed to be the current average European
Although the influence of both attributional and consequential production (for wood: round wood, for steel: a mix of pro-
approaches is explored in the EoL scenarios, we would call duction from primary and recycled steel) or, in one EoL
the overall study primarily consequential. This is because the scenario for steel frames, production from recycled steel only.
aim of the comparison is to support more robust decision In other EoL scenarios, the cut-off method is applied to
making. For this reason, we study the consequences of choos- allocate the environmental impacts between the roof construc-
ing one of two engineering alternatives, in other words, the tion and the by-products of the disposal process, which means
consequences of increased production of either alternative. that only impacts directly caused by a product are allocated to
The choice of a consequential approach is manifested primar- that product. This is a common allocation method in LCA of
ily in the choices of system boundaries: we exclude processes open recycling systems and thus a common method in LCAs
that are assumed to result in identical impacts for the com- of construction materials. In our case, impacts from
pared product systems. The use of average European data manufacturing, use and disposal of the roof construction are
could, on the other hand, be viewed as adhering to an attribu- allocated to it, and impacts from the subsequent life cycle
tional tradition, but this is suitable as we are studying the processes of the by-products (e.g. reprocessing and final dis-
average European consequence of choosing either of two posal of the recycled materials) are allocated to those process-
alternative products. In practice, the consequence of choosing es and thus cut-off from the studied system. This approach can
between alternative construction materials will depend on be seen as attributional, as it does not account for conse-
which European country (or even in which part of a country) quences in the surrounding market. It is the use of different
the materials are produced, used and disposed of, and in each approaches when modelling the EoL scenarios that enables
such case, the choice between average and marginal data may the second aim of the paper to be fulfilled: the comparison of
significantly influence the LCA results. It is not within the attributional and consequential modelling approaches.
scope of this study to explore such micro-level consequential
effects, and European average data is thus deemed the most
suitable data to use. Besides, some authors have suggested 3 System descriptions
that in markets constrained by regulation (e.g. national biodi-
versity conservation policies), the marginal supply should be The functional unit (FU) is defined as the internal roof con-
defined as the planned/predicted supply rather than the struction necessary to support a 90-m2 roof for an industrial
uninstalled technology with the lowest long-term marginal hall—a typical area supported by one glulam beam or one
cost (Schmidt et al. 2011). At the European level, even ex- steel frame. A simple and typical construction for an industrial
trapolation from current average supplies could thus be a hall was chosen: a single, sloping roof with a 1:10 inclination
better estimate for the marginal supply than the uninstalled and an aluminium roof cover. We assume a full service-life to
supply with lowest long-term marginal cost. Nonetheless, in be 50 years for both constructions.
some of the EoL scenarios, data on a single technology is used It is reasonable to assume that the choice of roof cover and
in what can be called “cornerstone” (Lundie et al. 2004; materials in other parts of the building does not vary depend-
Pesonen et al. 2000) or “explorative” (Börjesson et al. 2005) ing on the internal roof construction; therefore, these parts of
scenarios—these should not be seen as marginal technology the building are excluded from the study. Also, impacts from
scenarios but as a means of exploring the possible range of the construction phase is disregarded, as data in Björklund and
 Int J Life Cycle Assess

Tillman (1997) and Björklund et al. (1996) show identical Table 1 shows inventory data. For all processes, inventory
energy use for the installation of wood and steel frame data representative for European conditions is used.
structures, and energy use can be assumed to be the References and further descriptions of all inventory datasets
main driver for impacts from the installation phase. can be found in Table S1 of the Electronic Supplementary
Maintenance is normally not required for internal roof Material.
structures, and therefore we assume maintenance-free struc-
tures. Figure 1 shows a flowchart of the main processes of the 3.2 End-of-life scenarios
studied systems.
The EoL scenarios are described in detail in Table 2.
3.1 Process descriptions Assumptions regarding the following features of the EoL
processes are tested: energy source used in demolition, fuel
Due to a lack of up-to-date European data, inventory data on type used for transportation to the disposal site, means of
glulam beam manufacturing in the US Pacific Northwest is disposal and method for handling the allocation of environ-
used (Puettmann et al. 2013). Manufacturing includes drying mental impacts to by-products of the disposal. Two assump-
of green lumber, grading of lumber, end jointing the lumber tions regarding technology development are tested: no devel-
into longer laminations, face bonding the laminations together opment from today’s technologies and that today’s low-
with resin, finishing and fabrication. The inputs (except glue) impact technologies have become representative for the aver-
and emissions are allocated between beams and by-products age future technologies (for these scenarios, the prefix “green”
(shavings and trimmings) on a mass basis. We assume that no is used in the scenario abbreviations). In “green” scenarios,
surface treatment is applied after installation, as this is usually wind power is assumed to replace diesel as the energy source
not required for internal roof structures (personal communi- in demolition, EoL transportation is assumed to run on rape
cation with Mats Axelsson, SP Technical Research Institute of methyl ester (RME) biodiesel instead of today’s European
Sweden, February 2012). fleet average fuel (low-sulphur diesel), and in scenarios where
Manufacturing of the steel frame includes cutting, bending, substitution is applied, a low-impact technology is assumed to
welding and drilling of steel sheets (Björklund et al. 1996). be substituted. “Inc” and “Re” in the abbreviations refer to the
Often, the steel frame is coated with a primer and possibly also means of disposal (incineration or recycling), and “Cut” and
a top coating. However, in principal, a coating is not “Sub” refers to the method for handling allocation problems
needed for an interior steel frame; therefore, it is omitted in (cut-off or substitution).
this study, which is consistent with previous studies (e.g. The average technology in 50 years can of course have
Björklund et al. 1996). lower impacts than today’s low-impact technology, but to
At the end of the service life, glulam beams are usually account for this would require inventory data which is not
recovered and used for new applications, or incinerated, often available at present time. However, the “NoEoL” scenarios—
as fuel for district heating systems (Carling 2008; Björklund in which all EoL processes are excluded—can be seen as
and Tillman 1997). These two disposal methods are studied in “green extremes”: environmental impacts of future processes
the different EoL scenarios. The glue is assumed not to give are considerably lower than of today’s and thus ignored. These
rise to more hazardous emissions in the incineration than scenarios are also reasonable if it is argued that rapid reduction
wood does (Erlandsson 2007). Steel is typically recovered of current levels of environmental impact is urgent and im-
from building sites and used for production of recycled steel pacts occurring in 50 years should be subject to a high dis-
(Björklund et al. 1996), thus recycling is the disposal method count rate, which eliminates them from consideration. The
studied in the EoL scenarios. The EoL scenarios are further three assumptions regarding future technology—today’s av-
described in section 3.2. erage technology, today’s low-impact technology and fully

Fig. 1 Process flowchart for

glulam beam and steel frame life Material production
cycles. Processes in boxes with Alternative heat District
production - heating
dashed lines are modelled Transportation
differently in the various EoL +
scenarios Glulam beam/ Construction & Municipal
Trans- Demo- Trans-
steel frame maintenance incineration/
portation lition portation
manufacturing (excluded) recycling

Heat and electricity production Fuel or electricity production

Manufacturing: year 0 Use: year 0- 50 End of life: year 50

Int J Life Cycle Assess

Table 1 Inventory data for the glulam beam and steel frame product the atmosphere during forest growth and the subsequent stor-
systems
age of carbon in the glulam beam.
Processes Amount Unit For EP, the main difference from GWP can be seen when
comparing scenarios IncSub and GreenIncSub; more nitrify-
Reference flow: glulam beam 1,280 kg ing emissions are avoided if heat from municipal biowaste is
Glulam beam manufacturing substituted compared to heat from natural gas. POCP results
Sawn softwood timber, rough, kiln dried 959 kg are similar to EP results, except for that it matters less what
Sawn softwood timber, rough, green 322 kg technology is assumed to be avoided in the consequential
Liquefied petroleum gas 3.50 L glulam beam scenarios with incineration as the means of
Gasoline 46.2 L disposal (compare scenarios IncSub and GreenIncSub).
Diesel 0.95 L The method for calculation of water depletion potential
Glue (PVAc) 21.4 kg excludes natural rainfall as an input to production, which is
Natural gas 10.5 m3 seen clearly in the low water depletion potential for glulam
Electricity (continental European mix) 183 kWh beams. This is the simplest consequential approach—consid-
Wood fuel 49.7 kg ering rain as an input that is independent of production rate
Polyethylene (for packaging in transportation) 8.34 kg (see, e.g. Peters et al. 2010). The high water depletion poten-
Demolition, energy (diesel or wind) 225 MJ tial for the steel frames originates mainly from iron ore mining
Disposal by municipal incineration (by-product: 20,480 MJ for production of primary steel: compare steel frame scenarios
heat, substitutes natural gas or municipal biowaste) ReSub, in which we assume that the recycled steel substitutes
Disposal by recycling (by-product: wood, substitutes 1,280 kg
European average (mix of primary and recycled) steel pro-
debarked round wood)
Reference flow: steel frame 1,270 kg duction, and GreenReSub, in which we assume that it substi-
Steel frame manufacturing tutes steel production from recycled steel only. In one sense,
Electricity 2,540 MJ all scenarios are attributional in terms of the response of
Steel (low-alloyed) 1,270 kg hydrological systems to forestry. If one instead considers the
Demolition, energy (diesel or wind) 229 MJ consequences of forest harvesting on the hydrological behav-
Disposal by recycling (by product: steel, substitutes 1,270 kg iour of forest soils (e.g. increased runoff after harvesting—see
average low-alloyed steel or recycled un- and Sandin et al. 2013), one could estimate a hydrological conse-
low-alloyed steel) quential version of each of the scenarios, which would influ-
Transportation of input materials: 20–28 t lorry 200 km ence the impact assessment of water use.
(fleet average)
Transportation to building site: 20–28 t lorry 500 km Total PE is the indicator where glulam beam and steel
(fleet average) frame scenarios overall are most similar, while there are
Transportation to disposal site: 20–28 t lorry 50 km considerable differences in terms of non-renewable PE. The
(fleet average or RME) reason for the negative non-renewable PE score in glulam
beam scenario IncSub is the substitution of natural gas.
Furthermore, the high total PE of the glulam beam scenario
GreenIncSub is because the substituted energy source is a
disregarding any impact from future technology—can be seen municipal biowaste, which—as it is a waste product—is not
as a sensitivity analysis of the inventory data for the EoL allocated any PE during production.
processes.

5 Discussion and conclusions

4 Results
Despite our initial concern that the results of this comparative
Figure 2 shows the results of the impact assessment (numeric LCA might be confounded by different approaches in conse-
results can also be found in Table S2 of the Electronic quential modelling, and by differences between attributional
Supplementary Material). In all scenarios, the glulam beam and consequential modelling, the comparative analysis is re-
exhibits lower GWP, irrespective of whether an attributional, a markably robust. In relative terms (i.e. in terms of the ranking
consequential or an EoL exclusion approach is applied, al- between glulam beams and steel frames), assumptions in EoL
though magnitudes differ considerably. Green scenarios show modelling are of low importance. In comparable scenarios
notably higher GWP in the substitution scenarios; this is (i.e. scenarios with equivalent assumptions), glulam beams
because the substituted EoL treatment technology is cleaner, have clear environmental benefits compared to steel frames,
resulting in lower avoided impacts. The negative GWP in except for the scenario in which steel frames are recycled and
glulam manufacturing is due to the carbon withdrawn from today’s average steel production is substituted (ReSub) in
 Int J Life Cycle Assess

Table 2 Description of the EoL scenarios

Abbreviation Energy source Fuel in EoL Means of Method for handling the allocation Attr. (A)/
in demolition transportation disposal problems related to EoL processes Cons. (C)

Glulam beam scenarios

IncCut Diesel Average Incineration Cut-off A
IncSub Diesel Average Incineration Substitution of combustion of natural gas C
GreenIncCut Wind RME Incineration Cut-off A
GreenIncSub Wind RME Incineration Substitution of combustion of municipal biowaste C
ReCut Diesel Average Recycling Cut-off A
ReSub Diesel Average Recycling Substitution of today’s average European C
production of debarked round wood
GreenReCut Wind RME Recycling Cut-off A
GreenReSub Wind RME Recycling Substitution of today’s average European C
production of debarked round wooda
NoEoL All impacts of EOL processes are excluded
Steel frame scenarios
ReCut Diesel Average Recycling Cut-off A
ReSub Diesel Average Recycling Substitution of today’s average European production C
of low-alloyed steel
GreenReCut Wind RME Recycling Cut-off A
GreenReSub Wind RME Recycling Substitution of today’s average production of recycled C
un- and low-alloyed steel
NoEoL All impacts of EoL processes are excluded
a
This is the assumed substituted technology also for the “green” scenario, as we see it as a mature technology with limited development potential

which the impact is similar to the comparable glulam beam GWP scores would be somewhere between the scenario fully
scenarios. The choice of methodological approach (attribu- excluding EoL processes (NoEoL) and the other scenarios;
tional, consequential or fully disregarding EoL processes) thus the approach of building cornerstone scenarios does to
does not seem to influence the relative performance of the some extent capture the range of possible outcomes of tem-
compared roof construction elements. porally more dynamic impact assessments.
On the other hand, in absolute terms, assumptions in EoL By accounting for uncertainties in EoL modelling, the
modelling have a significant influence on the studied impact present study strengthens the results of other studies that have
categories. Four factors are particularly crucial: whether EoL also found potential environmental benefits of wood com-
phases are considered at all (compare NoEoL scenarios with pared to alternative construction materials (Bribián et al.
other scenarios), whether recycling or incineration is assumed 2011; Lippke et al. 2010; Upton et al. 2008; Petersen and
in the disposal of glulam beams, whether a consequential Solberg 2005). However, decisions for sustainable develop-
(substitution) or attributional (cut-off) approach is used for ment in the construction industry should not solely be based
modelling the disposal processes and whether today’s average on the impact categories discussed in the present paper but
technology or a low-impact technology is assumed for the also on the other impact categories recommended in EN
substituted technology (compare glulam beam scenarios 15804 (SIS 2012) and other impact categories of potential
IncSub with GreenIncSub or steel frame scenarios ReSub with importance for wooden materials, for example, related to
GreenReSub). In contrast, the assumed technologies of demo- toxicity or land use. The latter two impact categories are more
lition and transportation processes are of low importance supply-chain specific than the impacts presented here and are
(compare regular with green cut-off scenarios). thus difficult to account for in studies on generic European
To include a temporal dimension in the mapping of the products. Ultimately, decisions will likely depend on the pri-
product system, as done in the present study, offers the possi- orities given to different environmental concerns, as also
bility to use temporally more dynamic methods for the impact recognised elsewhere (Thiers and Peuportier 2012).
assessment of climate change (for review of methods, see As our results indicate that EoL assumptions can be highly
Brandão et al. 2013) and other impact categories. For exam- important for LCAs of construction materials—particularly in
ple, if emissions occurring during EoL processes would be absolute terms—we recommend that EoL uncertainties are
discounted—which under certain circumstances is allowed by taken into consideration in any LCA of long-lived products.
the ILCD Handbook (European Commission 2010)—the This could, for example, be done as in this study: by
Int J Life Cycle Assess

Manufacturing
Dotted bars = Glulam beam scenarios
EoL
Checkered bars = Steel frame scenarios
Net impact
EoL processes
Attributional EoL scenarios Consequential EoL scenarios excluded
3000
(kg CO2 eq./FU)
GWP

-3000
7.5
(kg PO43- eq./FU)

0
EP

-7.5
(kg ethylene eq./FU)

2
POCP

-2
potential (m3/FU)
Water depletion

20000

-20000

40000
Total PE
(MJ/FU)

-40000
Non-renewable PE

40000
(MJ/FU)

-40000

Fig. 2 LCA results for glulam beams and steel frames in different EoL potential, total primary energy demand and non-renewable primary ener-
scenarios in the impact categories of global warming potential, eutrophi- gy demand. See Table 2 for a description of the scenario abbreviations
cation potential, photo-chemical oxidation potential, water depletion

introducing a temporal dimension in the mapping of the future. We chose a 50-year time horizon based on the longev-
product system and using explorative scenarios to test as- ity of the studied constructions. Longer time frames are feasi-
sumptions for processes occurring in a distant and uncertain ble with the method, but the problem of technological
 Int J Life Cycle Assess

uncertainty becomes extreme. Furthermore, while straight- and scenario techniques. Environmental Strategies Research,
Department of Urban Studies, Royal Institute of Technology,
line depreciation times of 100 years do exist in financial
Stockholm
accounting for long-lived assets, the utility of an analysis that Bouhaya L, Le Roy R, Feraille-Fresnet A (2009) Simplified environmen-
is concerned with future issues typically discounted to zero by tal study on innovative bridge structures. Environ Sci Technol 43:
decision-makers today is questionable. 2066–2071
Brandão M, Levasseur A, Kirschbaum MUF, Weidema BP, Cowie
For the studied construction materials, EoL modelling
AL, Jørgensen SV et al (2013) Key issues and options in
should particularly focus on uncertainties regarding the means accounting for carbon sequestration and temporary storage in
of disposal, the expected technology development of disposal life cycle assessment and carbon footprinting. Int J Life Cycle
processes and the type of substituted technology. Also, the Assess 18(1):230–240
Brander M, Wylie C (2012) The use of substitution in attributional life
choice between attributional and consequential approaches
cycle assessment. Greenh Gas Meas Manag 1:161–166
should be treated carefully. It has been recommended earlier Bribián IZ, Capilla AV, Usón AA (2011) Life cycle assessment of
that several fundamentally different scenarios are needed building materials: Comparative analysis of energy and environ-
when modelling future disposal processes, particularly if a mental impacts and evaluation of the eco-efficiency improvement
potential. Build Environ 26:1133–1140
consequential approach with substitution is applied
BSI (2011) PAS 2050:11, Specification for the Assessment of the Life
(Mathiesen et al. 2009). Scenarios can also be used to capture Cycle Greenhouse Gas Emissions of Goods and Services. http://
the influence of other crucial uncertainties in LCAs of build- shop.bsigroup.com/en/forms/PASs/PAS-2050. Accessed 18
ing materials (e.g. as done by Cellura et al. 2011). February 2013
Carling O (2008) Limträ: handbok (English: “Glulam: handbook”). Print
More research is warranted to explore other attributional
and Media Center i Sundsvall AB, Sundsvall, Sweden. http://www.
allocation methods (e.g. based on physical or monetary prop- svensktlimtra.se/Upload/File/publikationer/2009/Limtrahandbok_
erties) in EoL modelling of long-lived products or to explore 200812.pdf. Accessed 7 March 2012
what EoL assumptions that are of importance for other con- Cellura M, Longo S, Mistretta M (2011) Sensitivity analysis to quantify
struction materials and other long-lived products (e.g. infra- uncertainty in life cycle assessment: the case study of an Italian tile.
Renew Sust Energ Rev 15(9):4697–4705
structure such as roads). The proposed research could further Cuéllar-Franca RM, Azapagic A (2012) Environmental impacts of the
clarify when and how EoL assumptions matter and thereby UK residential sector: life cycle assessment of houses. Build
contribute to more robust decision making in the construction Environ 54:86–99
sector and potentially also for civil engineering in general. Dixit MK, Fernández-Solís JL, Lavy S, Culp CH (2012) Need for en
embodied energy measurement protocol for buildings: a review
Until now, there has been insufficient consideration of EoL paper. Renew Sust Energ Rev 16:3730–3743
uncertainties of long-lived products, which may hamper Erlandsson M (2007) Miljödeklaration: limträ (English: “Environmental
sound decision making for sustainable development. product declaration: glulam”). http://www.svensktlimtra.se/Upload/
File/publikationer/Limtra_miljovarudeklaration%20.pdf. Accessed
6 February 2013
Acknowledgments This research was funded by the EU FP7 grant
European Commission (2010) International Reference Life Cycle Data
246434, WoodLife.
System (ILCD) Handbook—general guide for life cycle assessment
—detailed guidance, 1st edn. Publications Office of the European
Union, Luxembourg
References Frijia S, Guhathakurta S, Williams E (2011) Functional unit, technolog-
ical dynamics, and scaling properties for the life cycle of residencies.
Environ Sci Technol 46:1782–1788
Ardente F, Beccali M, Cellura M, Mistretta M (2008) Building energy Goedkoop M, Heijungs R, Huijbregts M, De Schryver A, Struijs J, Van
performance: a LCA case study of kenaf-fibres insulation board. Zelm R (2012) ReCiPe 2008 (first edition) – report I: characterisa-
Energy Build 40:1–10 tion (updated 13 July 2012). http://www.lcia-recipe.net. Accessed
Arvidsson R, Fransson K, Fröling M, Svanström M, Molander S (2012) January 21 2013
Energy use indicators in energy and life cycle assessments of Guinée JB, Gorrée M, Heijungs R, Huppes G, Kleijn R, Koning A et al
biofuels: review and recommendations. J Clean Prod 31:54–61 (2002) Handbook on life cycle assessment. Kluwer, Dordrecht
Beccali M, Cellura M, Fontana M, Longo S, Mistretta M (2013) Energy Habert G, Arribe D, Dehove T, Espinasse L, Le Roy R (2012) Reducing
retrofit of a single-family house: life cycle net energy saving and environmental impact by increasing the strength of concrete: quantifi-
environmental benefits. Renew Sust Energ Rev 27:283–293 cation of the improvement to concrete bridges. J Clean Prod 35:250–
Björklund T, Tillman A-M (1997) LCA of Building Frame Structures: 262
Environmental Impact over the Life Cycle of Wooden and Concrete Jungbluth N, Emmenegger MF, Dinkel F, Stettler C, Gabor D, Chudacoff
Frames. Technical Environmental Planning Report 1997:2, M et al (2007) Life Cycle Inventories of Bioenergy. Ecoinvent report
Chalmers University of Technology, Gothenburg, Sweden no. 17. Swiss Centre for Life Cycle Inventories, Dübendorf
Björklund T, Jönsson Å, Tillman A-M (1996) LCA of Building Frame Kounina A, Margni M, Bayart J-B et al (2013) Review of methods
Structures: Environmental Impact over the Life Cycle of Concrete addressing freshwater use in life cycle inventory and impact assess-
and Steel Frames. Technical Environmental Planning Report 1996: ment. Int J Life Cycle Assess 18(3):707–721
8, Chalmers University of Technology, Gothenburg, Sweden Lippke B, Wilson J, Meil J, Taylor A (2010) Characterising the impact of
Blengini GA (2009) Life cycle of buildings, demolition and recycling carbon stored in wood products. Wood Fiber Sci 42:5–14
potential: a case study in Turin, Italy. Build Environ 44:319–330 Lundie S, Peters G, Beavis P (2004) Life cycle assessment for sustainable
Börjesson L, Höjer M, Dreborg K-H, Ekvall T, Finnveden G (2005) metropolitan water systems planning—options for ecological sus-
Towards a user’s guide to scenarios—a report on scenario types tainability. Environ Sci Technol 38:3465–3473
Int J Life Cycle Assess

Mathiesen BV, Münster M, Fruergaard T (2009) Uncertainties related to Sandin G, Peters GM, Svanström M (2013) Moving down the cause-
the identification of the marginal technology in consequential life effect chain of water and land use impacts: an LCA case study of
cycle assessments. J Clean Prod 17:1331–1338 textile fibres. Resour Conserv Recy 73:104–113
Ortiz O, Pasqualino JC, Castells F (2010) The environmental Schmidt J H, Merciai S, Thrane M, Dalgaard R (2011) Inventory of country
impact of the construction phase: an application to composite specific electricity in LCA – Consequential and attributional scenari-
walls from a life cycle perspective. Resour Conserv Recycl os, Methodology report. 2.−0 LCA consultants. www.lca-net.com/
54:832–840 projects/electricity_in_lca. Accessed 5 October 2012
PE International (2013) GaBi software. http://www.gabi-software.com Singh A, Berghorn G, Joshi S, Syal M (2011) Review of life-cycle assess-
Persson C, Fröling M, Svanström M (2006) Life cycle assessment of the ment applications in building construction. J Arch Eng 17:15–23
district heat distribution system, part 3: use phase and overall dis- SIS (2012) SS-EN 15804:2012 Sustainability of construction works—
cussion. Int J Life Cycle Assess 11:437–446 environmental product declarations—core roles for the product
Pesonen HL, Ekvall T, Fleischer G, Huppes G, Jahn C, Klos SZ, Rebitzer category of construction products. Swedish Standards Institute,
G et al (2000) Framework for scenario development in LCA. Int J Stockholm
Life Cycle Assess 5:21–30 Thiers S, Peuportier B (2012) Energy and environmental assessment of
Peters GM (2009) Popularize or publish—growth in Australia. Int J Life two high energy performance residential buildings. Build Environ
Cycle Assess 14:503–507 51:276–284
Peters GM, Wiedemann SG, Rowley HV, Tucker RW (2010) Accounting Thormark C (2002) A low energy building in a life cycle—its embodied
for water use in Australian red meat production. Int J Life Cycle energy, energy need for operation and recycling potential. Build
Assess 15(3):311–320 Environ 37:429–435
Petersen A-K, Solberg B (2005) Environmental and economic impacts of Upton B, Miner R, Spinney M, Heath LS (2008) The greenhouse gas and
substitution between wood products and alternative materials: a energy impacts of using wood instead of alternatives in residential
review of micro-level analyses from Norway and Sweden. Forest construction in the United States. Biomass Bioenerg 32:1–10
Policy Econ 7:249–259 Verbeeck G, Hens H (2007) Life cycle optimization of extremely low
Puettmann M, Olein E, Johnson L (2013) Cradle to gate life cycle assessment energy dwellings. J Build Phys 31(2):143–178
of glue-laminated timbers production from the Pacific Northwest. http:// WBCSD/WRI (2011) Product Accounting & Reporting Standard. www.
www.corrim.org/pubs/reports/2013/phase1_updates/PNW% ghgprotocol.org/files/ghgp/Product%20Life%20Cycle%20
20Glulam%20LCA%20report%201_7_13%20final.pdf. Accessed 6 Accounting%20and%20Reporting%20Standard.pdf. Accessed 18
February 2013 February 2013