Comparative LCA of two approaches with different emphasis on energyor material recovery for a municipal solid waste management systemin Gipuzkoa

G. Bueno a,n, I. Latasa b, P.J. Lozano b

a Department of Electronics Engineering, Faculty of Engineering, University of the Basque Country UPV/EHU, Alameda Urquijo s/n, 48013 Bilbao, Spainb Department of Geography, Prehistory and Archaeology, Faculty of Arts, University of the Basque Country UPV/EHU, Tomás y Valiente s/n,01006 Vitoria-Gasteiz, Spain

a r t i c l e i n f o

Article history:Received 20 May 2014Received in revised form20 March 2015Accepted 1 June 2015

Keywords:Life cycle assessment (LCA)Municipal solid waste (MSW)Material recoveryEnergy recoveryWaste management

a b s t r a c t

Two alternative approaches for an integrated municipal solid waste management system (MSW-MS)have been confronted in the province of Gipuzkoa, in the north of Spain, during the last decade. Whileone of them prioritizes energy recovery from mixed residual waste in an incineration plant, the otherapproach gives precedence to material recovery of separately collected waste. Which system wouldpresent a lower environmental impact and be more desirable from a sustainability perspective?Answering this question is hindered by the fact that recovered energy and materials are not directlycomparable or directly substitutable with each other.

Based on the powerful framework provided by life cycle assessment (LCA) methodology, this workperforms a comparative LCA of overall environmental impacts of these two alternative approaches,showing that comparisons of alternative systems in terms of direct energy recovery or direct materialrecovery should be avoided in favor of other indicators already proposed in the LCA framework, such asthe Cumulative Energy Demand category from Ecoinvent, or the global warming potential and theAbiotic Resources Depletion categories from the CML 2001 method.

Applying the LCA framework, this work shows that when a high share of waste is collectedseparately, and processes assumed in the background system are adequately characterized, especiallythe production of the electricity mix, then prioritizing material recovery provides better results even inenvironmental categories tightly related to fossil energy consumption, such as the global warmingpotential impact category.

& 2015 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4501.1. Waste management strategies in Gipuzkoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4501.2. Objectives of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

2. Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4522.1. Goal and scope definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4522.2. Waste prevention derived from the broadening of selective collection in Gipuzkoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4532.3. Characterization of background and foreground processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/rser

Renewable and Sustainable Energy Reviews

http://dx.doi.org/10.1016/j.rser.2015.06.0211364-0321/& 2015 Elsevier Ltd. All rights reserved.

Abbreviations: acid, Acidification impact category from CML 2001 method; ard, Abiotic Resource Depletion impact category from CML 2001 method; eutro, Eutrophicationimpact category from CML 2001 method; GHG, Greenhouse Gas; gw, Global Warming impact category from CML 2001 method; htox, Human Toxicity impact category fromCML 2001 method; ILCD, International Reference Life Cycle Data System; ISO, International Organization for Standardization; LCA, life cycle assessment; LCA-IWM, LCA Toolsfor the Development of Integrated Waste Management; MBP, mechanical biological pre-treatment; MSW, municipal solid waste; MSW-MS, municipal solid wastemanagement systems; P, product; PE, primary energy demand; ph-tox, Photo-oxidant Formation impact category from CML 2001 method; RM, resource material demand;SC, separate collection; WFD, Waste Framework Directive; WP, waste prevention; WtE, Waste-to-Energy, incineration plant with energy recovery

n Corresponding author. Tel.: þ34 94 601 41 34; fax: þ34 94 601 42 59.E-mail addresses: gorka.bueno@ehu.es (G. Bueno), itxaro.latasa@ehu.es (I. Latasa), pedrojose.lozano@ehu.es (P.J. Lozano).

Renewable and Sustainable Energy Reviews 51 (2015) 449–459

3. Results and discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4544. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 456Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457Appendix A. System expansion to determine avoided burdens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458

1. Introduction

The aim of integrated municipal solid waste managementsystems (MSW-MS) is to give an adequate treatment to collectedwaste with a minimum environmental impact under affordablecosts. These systems comprise all the treatment and processingsteps underwent by collected fractions of municipal solid waste(MSW) generated in a specific area, from temporary storage andcollection through final disposal of secondary fluxes generated inprocessing plants. In order to improve sustainability and minimizeimpacts, some waste treatments—such as incineration or anaero-bic digestion—aim at recovering energy from waste, while othersare focused on preparing the waste for material recovery. In fact,integrated MSW-MS normally combine different kinds of materialand energy recovery.

1.1. Waste management strategies in Gipuzkoa

Local administrations in Spain have been redefining theirmunicipal waste-management systems for more than a decade.On one hand, they are obliged to comply with European Directivesregarding minimum recovery and recycling rates for packagingwastes and closure of landfills; on the other hand, many admin-istrations have to face up to the saturation of landfill sites. This isthe case, for example, in the Basque province of Gipuzkoa, where64% of all MSW generated in 2012 was derived to landfills. Thisfigure, actually, is similar to the values registered in nearbyprovinces and regions in Spain, as can be checked in Table 1,which shows the percentages of MSW derived to final treatmentsthat year in the three Basque provinces and Spain. There, treat-ment of MSW has been mainly based in landfilling and to a muchlesser degree in energy recovery; material recovery, on the otherhand, has remained below 40% for many years [1–4].

With a population of 731 thousand inhabitants in 2013,Gipuzkoa is administratively divided into eight municipality com-monwealths. Historically, municipality commonwealths are theadministrative bodies that have been in charge of the collectionand treatment of municipal waste, especially through its disposalto controlled landfills. Fig. 1 shows the trend of MSW generation inGipuzkoa between 2000 and 2013, altogether with planningobjectives established by the provincial administration in 2008(DdP-2008 Strategy, for year 2016 [5]) and in 2012 (EDDdP-2012revision Strategy [1], for 2016 and 2020).

MSW generation in Gipuzkoa increased since 2000 until 2006,when a peak of 411 thousand metric tons was generated. Duringthat period around 80% of the MSW was mixed residual wastesderived to landfills, as most of the waste was not separately

collected—from 15.3% in 2000 up to 25.5% in 2006. In order toreduce environmental impacts related to such a big waste fluxbeing derived to landfill sites, during those years the provincialadministration made a strong commitment to energy recovery ofthe mixed residual waste. This commitment was materialized inthe DdP-2008 Strategy, approved in the beginning of 2008. Thisplanning projected a progressive increase in waste generation andrecycling until 2016. According to it, in that year 57% of thegenerated waste would be separately collected and 53.3% couldbe recycled [5]. Most of the resting mixed residual waste (213thousand metric tons, annually) would be incinerated with energyrecovery. This strategy would have required the installation of atleast one new incineration plant in Gipuzkoa, although up to threenew plants were eventually considered [5,7]. It must be empha-sized that the DdP-2008 Strategy was established previous to theapproval of the European Waste Framework Directive (WFD),which sets a minimum target of 50% for re-use and recycling ofMSW by 2020 [8]. That target could be tightly achieved inside theDdP-2008 Strategy by 2016, but some serious problems arise whenthe evolution of MSW generation in Gipuzkoa after 2006 isconsidered.

Since 2007 the MSW flux generated in Gipuzkoa has dimin-ished steadily, as can be checked in Fig. 1. This reduction in wastegeneration seems to be due, partially at least, to a social contextmore sensible every year with recycling, re-use and environmentalimpacts derived from landfilling, as the decline started before theeconomy got into recession by the end of 2008. At that moment,MSW generation in Gipuzkoa had already diminished by 15%when compared to 2006 levels. By 2013 the reduction was 22%,and 35% less than the forecast for 2016.

After the approval of the DdP-2008 Strategy and the WFD in2008, some municipalities boosted an alternative approach inorder to avoid the installation of any new incineration facility inthe province. This alternative strategy was mainly based on a

Table 1Final treatments of MSW in 2012 in Gipuzkoa and nearby regions (other BasqueProvinces and Spain).

Final treatment Gipuzkoa (%) Bizkaia (%) Araba (%) Spain (%)

Landfilling 64 28 63 63Energy recovery 0 36 2 10Material recycling 29 36 34 17Composting 7 o1 1 10

Fig. 1. Historical evolution of the MSW flux in Gipuzkoa, and planning objectivesestablished by the DdP-2008 Strategy (for year 2016) and those established by theEDDdP-2012 revision Strategy (for years 2016 and 2020). Broken lines are eyeguides. Source: [1,5,6].

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459450

strong commitment to separate collection of household wastes,which would allow for the separate recovery of each materialfraction, and thus minimizing the need for final disposal tolandfills and incineration. A change in the provincial governmentin 2011 allowed a further implementation of this alternativeapproach. The new provincial government revised the DdP-2008Strategy in 2012, which materialized in an updated waste manage-ment planning for the period 2012–2016, the EDDdP-2012 revisionStrategy [1]. This updated planning took into account the newwaste generation trend after the 2006 peak, and reformulatedseparate collection and recycling targets for years 2016 and 2020,improving the targets imposed by the WFD: by the end of thedecade 76% of MSW generated in Gipuzkoa would be separatelycollected, which could boost materials recycling well over 70%. Inthe new planning, by 2020 the residual fraction would be reduceddown to 77 thousand metric tons annually, or 36% of the flux thatin the previous planning was supposed to be needed to feed thenew incineration facility, 213 metric tons. Under these circum-stances of more ambitious recycling targets and less MSW gen-eration the economical viability of the incineration facility wouldbe seriously jeopardized, as its functioning would diverge toomuch from full capacity [9].

In the context of this socio-political debate—not exempt ofunderstandable economic conflicts, as waste managementdemands a significant part of every municipal budget, even intimes of economic turndown—, social agents and decision-makersfrom Gipuzkoa have addressed our research group with questionssuch as the following, to be answered from a technical andscientific point of view: Which kind of recovery has to be givenprecedence in a waste-management system—energy or materialrecovery? Which is the significance of separate collection in anintegrated MSW-MS such as the one to be implemented inGipuzkoa?

1.2. Objectives of the study

The framework necessary to answer those previous questions isalready settled in the WFD, which establishes, through its wastehierarchy, a legally binding priority order for waste managementin the EU [8]. Prevention and preparing for re-use rank at the topof the hierarchy, followed by different kinds of material andenergy recovery. This hierarchy is not arbitrary, as the WFD statesthat potential deviations from it—and the choice among alterna-tives at each hierarchy level—have to be justified by life cyclethinking of the overall impacts. This is often achieved by theapplication of life cycle assessment (LCA), which is a preferred andstandardized scientific approach for life cycle thinking. The basicframework for LCA is provided by the ISO 14040 and 14044:2006standards [10,11]. Handbooks are available for its application [12],along with an international reference guide [13] and a guidancefor its application in waste management [14], where a number ofmodels have been developed during the last two decades [15]. Theuse of these models abounds in the literature, and they areespecially suited for the assessment of integrated MSW-MS thatmay combine energy and material recovery from waste. De Feoand Malvano [16] use the WISARD LCA tool in selecting the bestMSW-MS for the Campania Region, in Southern Italy. Bovea et al.[17] make use of SimaPro7 for the assessment of alternatives in theSpanish town of Castellón de la Plana. Pire et al. [18] carry out anLCA for a future MSW-MS in the Setúbal peninsula, in thePortuguese region of Lisbon, using the Umberto 5.5 software.Tunesi [19] uses the WRATE modeling tool for the assessment ofdifferent energy recovery strategies in England. Slagstad andBrattebø [20] use EASEWASTE to assess different alternatives forwaste management in a new urban settlement in the city ofTrondheim, in central Norway. Song et al. [21] use SimaPro7 for

the assessment of environmental performance of MSW-MS inMacau, China. Bernstad and la Cour Jansen [22] compare differentalternatives for the integrated management of household foodwaste in the area of Augustenborg, Southern Sweden, using theEASEWASTE LCA-tool. Eriksson et al. [23] study different MSW-MSfor the Swedish municipalities of Uppsala, Stockholm and Älvda-len, using the ORWARE model. Merrild et al. [24] assess recyclingversus incineration in waste management systems in Denmark, bymodeling in EASYWASTE. Nadzirah Othman et al. [25] review sixlife cycle assessments of integrated MSW-MS in Asian countriesthat combine both energy and material recovery approaches.

The main objective in this work is to determine whichintegrated MSW-MS may cause in the province of Gipuzkoa alower environmental impact and be more desirable from asustainability perspective—either a management system thatprioritizes energy recovery from mixed residual waste in anincineration facility, or another one that gives precedence tomaterial recovery of separately collected waste. In order tocompare these two alternative approaches, this work carries outa comparative LCA of these two alternatives, to be implemented ina generic municipality commonwealth. The modeling of thisgeneric municipality commonwealth is based on the presentcontext of Gipuzkoa, and its detailed characterization is performedin the following Section 2. We believe that the quantitativeassessment of environmental impact indicators in a genericmunicipality commonwealth allows drawing some importantqualitative conclusions that may be valid not only for the wholeprovince of Gipuzkoa, but also for other provinces or regions witha similar socio-economic situation and waste treatment conditionsin Spain, as shown in Table 1.

Some methodological choices may have important conse-quences when performing a comparative LCA of alternativewaste-management systems. Gentil et al. [15] reviewed theimportance of technical assumptions related to the definition ofthe functional unit, system boundaries, and energy and processmodeling in LCA models, concluding that making different choicesmay lead to contradictory results. Other important factors mayalso have important effects when assessing the environmentalimpact of waste-management systems, such as considering differ-ent waste prevention strategies, different collection systems, ordifferent spreading levels of separate collection. Regarding towaste prevention (WP), Gentil et al. [26] evaluated several mea-sures for municipal waste management; Slagstad and Brattebø[20], on the other hand, quantified WP potential to reduce house-hold waste generation in circa 17% for a new urban settlement inNorway. Other studies have centered on the influence of differentcollection systems, altogether with different treatment options[17,20,27]. The spreading of separate collection is also analyzed insome comparative LCA studies [16,28–32], but with quite differentranges under consideration: while Buttol et al. [29] assumed verylimited variations in separate collection, Rigamonti et al. [30]considered a range from 35% up to 60%, Calabrò [31] from 15%up to 50%, Consonni et al. [32] from 35% to 65%, and De Feo andMalvano [16] from 35% up to 80%. But other studies do notconsider any increase in separate collection, e.g. Cimpan andWenzel [33] and Belboom [34] when comparing different pre-treatments of residual waste, or Koci and Trecakova [35] whencomparing different treatments of mixed residual waste. Similarly,the possibility to increase separate collection is absent in otherstudies that compare different technologies for incineration[36,37], that compare final disposal to landfill versus incineration[38], different ways for energy recovery [39,40], or that comparematerial versus energy recovery [24].

Taking all this into account, it is also an objective of this work tocheck the importance of the spreading of separate collectionof MSW on the overall environmental balance of integrated

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459 451

MSW-MS, along with other factors such as the presence of wasteprevention strategies, and the adequate characterization of theelectricity mix generation in the background process.

This work also aims to demonstrate that the LCA methodologyframework provides a set of indicators, such as the CumulativeEnergy Demand category from Ecoinvent, or the global warmingpotential and the Abiotic Resources Depletion categories from theCML 2001 method, that allow to assess and compare life-cyclematerial and energetic consumption in systems of very differentnature that involve energy fluxes and material resources notdirectly comparable or directly substitutable with each other.

2. Materials and methods

2.1. Goal and scope definition

As this study is centered in the proper accounting of differentenvironmental impacts when comparing systems, the attribu-tional modeling principle has been chosen for this comparativeLCA, and the system expansion/substitution approach has beenconsidered for solving multifunctionality (Situation C1 in [13]).

The comparative LCA is carried out with the LCA-IWM tool [41].The assessment tool of LCA-IWM allows comparing differentscenarios, based on the LCA methodology, considering all wastemanagement steps, from temporary storage through final disposalof secondary fluxes generated in previous treatments, such asrecycling, incineration or composting. This tool was speciallydesigned for planning and optimizing waste-management systemsin areas that still require much effort to be adjusted to the state-of-the-art in Europe, as is the case in Southern European countries,and particularly in Spain.

The general diagrams of the two integrated MSW-MS modeledwith the LCA-IWM assessment tool in this work are shownsuperimposed in Fig. 2, with the corresponding divergencesbetween them in fluxes and processing steps.

Our model considers five different waste flows separately col-lected: biowaste, glass, metals, plastics, and paper and cardboard, with

the specific compositions assumed in the LCA-IWM tool by default—for every parameter not specified from now on, LCA-IWM default datashould be assumed. The percentages of separately collected fractionsare specified in Table 2, and resemble those of Gipuzkoa in 2011 [1]. Asixth primary flow corresponds to the residual waste collected inmixed form, of which almost 70% is biowaste [42]. One of the keydifferences between the two systems considered affects the treatmentof this residual flow. On one hand, in the system that prioritizesmaterial recovery, this mixed residual flow is transported to anaerobic mechanical biological pre-treatment (MBP), where the organicfraction is stabilized, the high caloric fraction is recovered for itscombustion in cement kilns, and the resulting secondary residualwaste is left ready for its safe disposal to landfill. On the other hand, inthe alternative that prioritizes energy recovery, the residual flow isdirected to an incineration plant, and the ashes and the slag thereproduced are also landfilled, as the Basque legislation does not allowfor its use as gravel for road construction or similar.

Historically, MSW management systems in Spain have beenreliant on the disposal to landfills of not separately collected mixedwastes. In 2006, as much as 80% of household wastes in Gipuzkoawere collected this way [5]. In the nearby province of Bizkaia asimilar percentage was reached in 2013 [2]. In parallel, it is wellknown that small sized incineration plants are seriously handi-capped because of lower electric efficiencies due to scale effects,higher specific consumption of auxiliaries, and more conservativedesign conditions and less sophisticated configurations, as eco-nomic constraints are tighter in them [43]. Incineration plantsperform better if incoming waste fluxes are bigger. As theynormally recover energy from mixed wastes that cannot berecycled, administrations do not find much incentive to broadenselective collection schemes that reduce incoming waste fluxes toincineration plants and may jeopardize their viability. This isspecially the case in Gipuzkoa, where annual household wastegeneration barely exceeds 300.000 metric tons. On the contrary,systems that prioritize material recycling should always try toextend separate collection schemes, as only separately collectedwaste can be most satisfactorily recycled. Coherent with thisreasoning, our modeling assumes different separate collectionlevels for each system: 25% in the system with the incinerationplant, and 75% in the system with the aerobic MBP.

The two alternative MSW-MS analyzed in this work giveservice to a population of 100,000 inhabitants living in 25,000households in an area of 1000 km2 and generating an annualwaste flux of 50,000 metric tons when no waste preventionstrategies are put into action. These and other characteristics ofthe functional unit are gathered in Table 3.

At this point, an adequate definition of the functional unit iscrucial. Several problems related to the definition of the functionalunit arise when performing a comparative LCA of structurallydifferent waste-management systems.

Fig. 2. Material flux diagram of the two integrated MSW-MS considered inthis work.

Table 2Waste fractions considered in the functional unit.

Waste fraction System withincineration (%)

System withaerobic MBP (%)

Mixed residual waste, of which70% is bioresidue

75 25

Separately collected waste, ofwhich:

25 75

Paper and cardboard 24 24Glass 11 11Metals 5 5Plastics 15 15Biowaste 45 45

Total 100 100

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459452

One of these problems is to solve the allocation of impacts andbenefits of different systems that are intrinsically multifunctional,while maintaining the comparability of the systems through acommon functional unit to all of them. Along with the wastemanagement service, these integrated systems allow for therecovery of different recycled materials and energy carriers. Butas these recoveries are complementary to the waste managementservice, which is the common function to all systems, the func-tional unit of the systems compared in this work is defined as aservice: the collection and treatment of all household waste in thedefined area in one year. Once the functional unit is defined thisway, the multifunctionality problem can be solved by systemexpansion/subtraction. This process is thoroughly explainedin Appendix A.

2.2. Waste prevention derived from the broadening of selectivecollection in Gipuzkoa

Recent experience in several municipalities of Gipuzkoa showsthat the substitution of kerbside collection of mixed residual wasteby door to door collection of the different fractions—including avery small residual fraction—may significantly reduce the total fluxof the waste to be managed by the system. This is the case, forexample, of Hernani, a town of 19,300 inhabitants where theimplantation of door to door collection altogether with thepromotion of home and district composting and campaigns toraise public awareness has led to a stable reduction of 28.6% intotal generated municipal solid waste (Fig. 3; [46–48]).

This work compares two management systems with differentlevels of separate collection (SC), and thus that implement wasteprevention strategies up to different levels. This would be an exampleof waste prevention as a result of different system dynamics [49].

If the functional unit of the systems under comparison is definedas the one that provides the service for collection and treatment of allhousehold waste in a given area and year, then comparability ofdifferent waste-management systems is guaranteed only as long asprevented waste generation remains equal in all systems. Otherwise,the comparison must account for the avoided impacts in thosesystems that prevent more waste generation. Ways to solve thisproblem have been proposed [36,50,51]. Basically, these workspropose to consider the managed waste flux as the sum of thecollected and treated wastes plus a virtual flux corresponding to theprevented waste. The burdens associated to the prevented wasteshould be accounted, in that case, as avoided burdens of the waste-management system because of waste prevention. But this approachis not exempt from problems [26]. It requires the quantification of adematerialized flux [50], and entails abandoning the “zero-burden

assumption”, as upstream burdens carried about by prevented anddematerialized waste should be accounted. As this approach compli-cates significantly our comparative LCA, this work does withoutconsidering any virtual flux associated with prevented waste, butalways keeping in mind that an accounting error of avoided burdensis being committed in favor of those systems with less ambitiousprevention strategies.

2.3. Characterization of background and foreground processes

Electricity produced fromwaste, e.g. in incineration plants withenergy recovery, is credited in our comparative LCA with thecorresponding avoided burdens from power generation in thebackground system. Thus, electricity generation may cause a hugeimpact on the net environmental balance of the waste-management system. When crediting these avoided burdens,comparative LCAs in the literature often consider national andlocal electricity mixes with a very high penetration of fossil fuels[29,35,37,38]. In some studies the electricity mix of the back-ground system is not even characterized much farther than asstrongly based on fossil, and thus giving way to important avoidedburdens [52].

But LCA is often applied to systems that are being projected forthe near future [29,36,53] or although already functioning, that arenot expected to be dismantled soon [23]. If Attributional LCA isapplied for the modeling of future systems [54], it has to take into

Table 3Characteristics of the functional unit, and of processes that diverge from default options in the LCA-IWM assessment tool.

Data input to the LCA-IWM assessment tool

Population 100,000 inhabitantsArea 1000 km2

Number of households 25,000Waste generation 50,000 metric tons/yearReduction due to waste prevention No waste prevention (0%); 20%Temporary storage

Recycled materials 80 L sacksMixed residual waste 1100 L plastic bins

Collection and transportRecycled materials 150 days/year (biowaste)

100 days/year (others)Mixed residual waste 310 days/year (as in Bilbao [44] or Donostia [45])Fist pick-up distance 7.5 kmAverage distance from sector to facilities 10 km

Efficiency of incineration plant 25%Electricity mix profiles considered 211 g CO2/kW h (high penetration of renewables); 498 g CO2/kW h (mainly fossil generation)

Fig. 3. Evolution of municipal solid waste generated in Hernani (19,300 inhabi-tants, Gipuzkoa) in 2009, 2010 and 2011 before and after the implantation of doorto door collection in May 2010.

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459 453

account data from background processes as they are forecast to bein the future, when the system under study is supposed to be putinto operation. In our study the new incineration plant in one ofthe alternative systems would start operation not before 2015, andwould not finish its pay-off period until 2030 [9], being probablyin operation by the middle of the century. Taking into account thatthe European Commission plans that, due to fossil energy deple-tion and fight against climate change, the European power sectorshould reduce its GHG emissions between 54% and 68% in 2030and between 93% and 99% by 2050 [55], the average production ofelectricity to be considered in the background system cannot becarbon intensive.

Actually, Spain has already reduced its electricity mix emissionslevel from 430 g CO2/kWh in 2000 [56] down 236 g CO2/kWh in2013 [57], and will probably reduce it further during the nextdecade, well below 200 g CO2/kWh. Following this trend, ourcomparative LCA will consider for the background system anemissions level of 211 g CO2/kWh, corresponding to an electricitymix with a high penetration of renewables. In order to perform asensitivity analysis of these avoided burdens, our study will alsoconsider another electricity mix, much more dependent on fossilfuels, with an emissions level of 498 g CO2/kWh. These twoelectricity mixes are characterized in the Ecoinvent-2000 database[58] and can be used by the LCA-IWM assessment tool.

The need to correctly address the average process is alsoapplicable to products obtained from material recovery. Whenassessing forecast systems, the LCA practitioner should also takeinto account that the production technologies of paper, plastics,ferrous metals, aluminum and organic fertilizers—which are dis-placed by compost—will probably reduce their burdens in thefuture, e.g. as it has occurred with the production of nitrogenousfertilizers, where using best available techniques may significantlyreduce N2O emissions and energy demand [22].

Also, sufficient information has to be provided about the assessedprocesses for energy and material recovery. In the case of ourcomparative LCA, these processes are those modeled by the LCA-IWM assessment tool, and characterized in its documentation [59]:

� The incineration plant is equipped with grate firing and flue gascleaning (electrostatic precipitator for dust and fly ashes; acidflue gas scrubbing for removal of HCl, HF and heavy metals;neutral SO2-scrubbing facility with suspended Ca(OH)2; filterswith activated carbon for removal of dioxines/furanes; andSelective Catalytic Reduction for denitrification). The Waste-to-Energy plant (WtE) produces only electricity, as climatic con-ditions in Gipuzkoa would not guarantee sufficient heatdemand from a CHP plant [60]. A thermoelectric efficiency of25% has been supposed, so that the incineration plant reachesthe R1 status of the WFD [61].

� For the recycling of plastics, it is assumed that plastics andcomposites separately collected are composed by the followingseven fractions: HDPE, PET, LDPE film, mixed plastics, liquidbeverage cartons, other composites, and contaminants (11%).These fractions are sorted in a Material Recovery Facility, andtransported to recycling facilities. Recycled HDPE substitutesprimary HDPE for multi-layered bottles (1:1 basis). RecycledPET substitutes primary PET for three-layered bottles (1:1).Recycled LDPE film substitutes primary LDPE for sacs (1:1).Mixed plastics are recycled into plastic pickets, which replacewood pickets (1:1 basis). Liquid beverage cartons are recycledinto pulp that substitutes primary pulp for domestic paper(1:1). Rejects of sorting processes and some composites areincinerated if the system has an incineration plant, otherwisethey are landfilled.

� Recycling of metals. To reprocess steel from scrap, first it issorted to remove contaminants, so that it can be melted and

recast. Tinplate is electrolytically de-tined to produce steel.Reprocessing of aluminum, which is much less energy intensivethan its production from virgin materials, requires sorting andthen melting in a furnace. Our model assumes that metals aresorted in a Material Recovery Facility and transported torecycling facilities, where tinplate steel is recycled into second-ary steel, substituting primary steel in a 1:1 basis; aluminum isrecycled into secondary aluminum, which substitutes primaryaluminum in a 1:1 basis. Rejects of sorting processes (5%) arelandfilled or incinerated.

� Related to recycling of paper and cardboard, following LCA-IWM, our model assumes that 1 kg of recycled pulp replaces1 kg of primary pulp, and that cardboard is recycled intocardboard. 2% rejects are derived to incineration if available;otherwise they are landfilled.

� Different subfractions of glass (green, brown, clear, mixed glass)are cleaned and crushed into broken glass in a MaterialRecovery Facility and transported to a recycling facility. Rejects(3%) of cleaning and crushing processes are landfilled orincinerated. Clean broken glass is recycled into glass, assumingthat 1 kg replaces 1.19 kg of raw materials.

� The modeled landfill is equipped with gas and leachate collec-tion systems. The collected gas is utilized for energy produc-tion, and leachate is treated before discharge.

� The composting process of the biowaste is modeled by the LCA-IWM tool assuming the operation of a fully encapsulatedcomposting plant with a first stage of intensive composting ina box system, and a subsequent maturation step in enclosedwindrows. Obtaining high quality compost is not a problemwhen the biowaste is separately collected. Its application bringspositive effects in form of nutrient and organic carbon supply,along with carbon sequestration. Our modeling assumesdefault parameters from the LCA-IWM tool, which imply thesubstitution of mineral fertilizers in a 1:1 basis (based on thenutrient content), the substitution of peat—which is considereda fossil resource—for introduction of organic matter to the soil,and carbon sequestration equivalent to 8.2% of the carbonpresent in final compost.

3. Results and discussion

In this section we present the results of the comparative life-cycle assessment of the two alternative integrated MSW-MSwhose characteristics have been previously detailed. These resultsare gathered in Table 4. The scenario labeled as A25 models thesystem in which 25% of waste is separately collected and the other75% of mixed residual waste is treated in a WtE plant. The scenariolabeled as B75 models the system in which 75% of waste isseparately collected, and the other 25% of mixed residual wasteis subjected to aerobic mechanical biological pretreatment andsubsequent disposal of nonrecyclable inert materials to landfill.Scenarios A25 and B75 are modeled assuming a power system inthe background with a high penetration of renewables (emissionslevel of 211 g CO2/kWh).

These two basic scenarios are complemented with other threein which some of the simulation conditions are modified in orderto perform sensitivity analysis of some significant parameters:

� In order to check the relevance of waste prevention andrecycling derived from the increase of selective collection,scenario B25 resembles scenario B75 but where just 25% ofwaste is separately collected, and there is no reduction in wastegeneration due to prevention.

� In order to check the relevance of the electricity mix assumedin the background, A25C and B25C scenarios model the

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459454

systems considered in scenarios A25 and B25, but assuming apower system in the background that is carbon intensive (498 gCO2/kWh).

Table 4 shows the five scenarios analyzed, with the parametersthat differentiate each one, and their modeling results for sixsignificant impact categories. These categories are those assessedby the LCA-IWM tool following the CML 2001 method [12], andthey are identified as the most significant when comparing waste-management systems. The first two, abiotic resource depletion(ard, measured in Mg Sb eq) and global warming potential (gw,measured in Gg CO2 eq) are very good indicators of cumulativematerial resource consumption (ard) and cumulative fossil energydemand (gw), representing very good indicators of global energyand material recovery. The other four impact categories analyzedare human toxicity (htox, measured in kg 1,4-Dichlorobenzene-eq),photo-oxidant formation (ph-tox, measured in kg Ethene-eq),acidification (acid, measured in kg SO2 eq) and eutrophication(eutro, measured in kg PO4 eq). Quantities of annual waste derivedto landfills are also gathered in Table 4 for each scenario, measuredin metric tons.

The first pair of scenarios shown in Table 4 (scenarios A25C-B25C) compare impact categories in both waste-managementsystems when separate collection is 25%, and a carbon intensiveelectricity mix is assumed in the background. The life cycleassessment provides better results (more negative) in all impactcategories for scenario A25C, showing that it is environmentallymore beneficial to incinerate the mixed residual waste than toinertize and dispose of it to landfill when just 25% of all generatedhousehold waste is separately collected.

The Spanish power sector is undergoing a decarbonizationprocess that will strengthen in the coming decades. Hence itseems more adequate to assume an electricity mix for the back-ground system less reliant on fossil fuels than that considered inscenarios A25C-B25C. The second pair of scenarios compared inTable 4 (A25C-A25) allows a sensitivity analysis of the electricitymix in the background. The comparison shows the consequence ofreducing the electricity emissions from 498 down to 211 g CO2/kW h in the systemwith the WtE plant: all environmental impactsremain beneficial due to important avoided burdens, but they aresignificantly reduced, from 32% (htox) up to 50% (gw).

Another factor that has to be considered when comparing thetwo alternative integrated MSW-MS is the possibility to increaseseparate collection. Rigamonti et al. [30] state that the optimumshare for separate collection may be around 50% due to contam-inations; but assuring high efficiencies in the separate collection ofeach fraction would locate the optimum well over 60%. Actually,

Slagstad and Brattebø [20] consider in their comparative assess-ment for a new urban settlement a feasible sorting efficiency of70% for food waste, and between 70% and 90% for all other wastefluxes. In our case, the third pair of scenarios compared in Table 4(B25-B75) perform a sensitivity analysis of the spreading ofseparate collection, comparing impact categories when it is 25%and 75% in the management system that derives the mixedresidual waste to aerobic MBP. The results show importantimprovements in all impact categories. This is due to the increasedavoided burdens that are accounted when tripling separate collec-tion, and thus material recovery. The improvement is significanteven in the global warming potential category, directly linked tofossil energy consumption (increase of 156%). It has to be addedthat this modeling underestimates the environmental benefit ofincreasing separate collection, as our modeling does not assignavoided burdens to a waste prevention that is estimated in 20%.

Direct energy recovery from waste is an environmentalimprovement when performed in a waste-management system.But the expansion of separate collection schemes provides envir-onmental benefits through expanded material recovery that mayoverwhelm those derived from energy recovery. A better resultfrom direct material recovery (e.g. recycling) when compared withdirect energy recovery (e.g. incineration) is confirmed by otherworks [27,62,63], and supports the fact that the former is locatedhigher in the waste hierarchy [8]. This point is confirmed by thelast pair of scenarios compared in Table 4 (A25-B75), where thewaste-management system with an incineration plant that sepa-rately collects just 25% of all household waste is compared withthe system that separately collects 75% for material recovery, andderives to aerobic MBP the mixed residual waste. This secondsystem (scenario B75) behaves better in all environmental cate-gories except eutrophication, in which the gap between the twosystems is nevertheless significantly reduced with respect toresults when separate collection is 25% in both systems (A25C-B25C).

Giving priority to material recycling over direct energy recov-ery improves material recovery, and therefore scenario B75 showsa better environmental impact in the Abiotic resource depletioncategory (–85.7 Gg Sb eq) than scenario A25 (–54.1 Gg Sb eq). Butresults show that overall energy recovery is also improved whenmaterial recovery is prioritized: scenario B75 shows a better resultin the global warming potential category (–11.09 Gg CO2 eq),closely related to fossil fuels consumption, than scenario A25(–4.76 Gg CO2 eq). This is due to the fact that important quantitiesof energy are required to produce materials that can be substitutedby recycled products. This energy consumption is avoided withmaterial recovery, and actually exceeds direct energy recovery

Table 4Parameter characterization and results of significant impact categories for five scenarios analyzed (A25, A25C, B25, B25C, B75), organized in four comparative pairs (A25C-B25C, A25C-A25, B25-B75, A25-B75) with the changing parameters in each pair in bold type.

Scenario Mixedresidualwastetreatment

Separatecollection(%)

Reductiondue towasteprevention(WP, %)

Electricitymix (gCO2/kW h)

Abioticresourcedepletion(ard, Mg Sb-eq)

Globalwarmingpotential(gw, Gg CO2

eq)

Human toxicity(htox, kg 1,4-Dichlorobenzene-eq)

Photo-oxidantformation(ph-tox kgEthene-eq)

Acidification(acid, kg SO2

eq)

Eutrophication(eutro, kg PO4

eq)

Wastelandfilled(tonnes)

A25C Incineration 25 No 498 –88.7 –9.56 –2.34 –4.76 –105 –946 7790B25C Aerobic

MBP25 No 498 –50.2 –5.04 1.44 –2.04 –56.8 2,352 27,175

A25C Incineration 25 No 498 –88.7 –9.56 –2.34 –4.76 –105 –946 7790A25 Incineration 25 No 211 –54.1 –4.76 –1.60 –3.15 –64.5 353 7790B25 Aerobic MBP 25 No 211 –45.0 –4.32 1.55 –1.80 –50.7 2,549 27,175B75 Aerobic MBP 75 Yes, 20% 211 –85.7 –11.09 –1.79 –6.55 –139 907 9939A25 Incineration 25 No 211 –54.1 –4.76 –1.60 –3.15 –64.5 353 7790B75 Aerobic

MBP75 Yes, 20% 211 –85.7 –11.09 –1.79 –6.55 –139 907 9939

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459 455

form waste in the considered systems. This is shown in Fig. 4,which details the partial contribution of each management stageand treatment process to the net environmental impact in scenar-ios A25 and B75.

Fig. 4 shows the importance of the avoided burdens in materialrecovery from the separately collected plastics, paper, glass andmetals residues. The avoided burdens are especially important formaterial recovery from plastics residues in the categories of abioticresource depletion and eutrophication; for recovery from glass inhuman toxicity; and for recovery from paper in photo-oxidantformation and acidification. Avoided burdens due to recovery frommetals seem to be less important in the category of humantoxicity, but are comparatively significant in all other categories.

Credits for the avoided burdens in material recovery are alsoimportant in the system with incineration, but these are lesssignificant than in the modeled system with aerobic MBP of themixed residual waste. Actually, most of the credits come from therecovery of materials separately collected, and therefore they keepapproximately proportional to the share of separate collection intotal waste collection. The increase of avoided burdens carried outby the increase of the share of separate collection in one system(B75) more than compensates for the credits gained in the othersystem when those residues are incinerated as part of the mixedresidual fraction (A25). Those credits, besides, are limited to theabiotic resource depletion and human toxicity categories, and tothe avoided burdens from the aerobic MBP—inexistent in thesystem with WtE plant—and also limited to the impact categoriesof human toxicity and eutrophication.

Composting biowaste provides some significant environmentalcredits, especially in the categories of global warming and humantoxicity. Inasmuch as composting of biowaste is not free of someemissions, especially of ammonia [64], those reflect with a significantimpact in the category of eutrophication, and with amuch lesser extentin the categories of photo-oxidant formation and acidification. Com-posting brings about with it some environmental impacts that wouldbe inexistent in a management system where most of the biowaste isincinerated. Nevertheless, assessment tools do not normally considersome environmental benefits of composting e.g. improvement of soilhealth, fertility and water retention capacity, and reduced pesticideconsumption [14]. In addition, other alternatives to the aerobic

processing of biowaste to produce compost could be also consideredas alternatives to biowaste incineration, such as anaerobic digestion,which, besides, allows for the direct recovery of energy by means ofbiogas production, along with other material recoveries (digestate). Theconsideration of these alternatives falls out of the scope of this paper,but other studies have already addressed a more beneficial net balanceof anaerobic treatments when compared with composting [27]. Never-theless, composting is credited as a very suitable biowaste treatmentoption for European Southern regions [65].

Another important environmental impact of the waste-management systems under analysis is the disposal to landfill offinal waste fluxes, mainly rejected materials in recycling plants,and slag and ashes from incineration. Although these secondarywastes generated in incineration plants are not statisticallyreported as part of the municipal waste data collected in Europe[66], in many countries landfilling is inseparable from incinerationif the complete life-cycle of municipal wastes is considered. This iswell known, for example, in land-scarce and incineration-intensive Singapore, where the spread of separate collection ofmunicipal waste is addressed as a key approach to reduce the needof almost saturated landfills for the disposal of slag and ashesgenerated in incineration plants [67].

Final waste fluxes disposed of to landfill in each scenario aregathered in the last column in Table 4. While the system withincineration and 25% of separate collection (scenarios A25, A25C)manages annually 50,000 metric tons of waste and derives tolandfill 7790 metric tons, the system without incineration underthe same conditions for separate collection (B25) derives to landfill27,175 metric tons of final residues. From this comparison we mayconclude that incineration is a viable strategy to reduce the flux offinal waste derived to landfill; but not the only strategy. Whenwaste prevention and the spreading of separate collection areimplemented in our model, the system without incineration(scenario B75) derives just 9939 metric tons to landfill, whichsupposes a reduction of 63.4%.

4. Conclusions

This work performs a comparative analysis of two alternativeapproaches for an integrated MSW-MS to be implemented in theBasque province of Gipuzkoa (Spain). These alternatives place differ-ent emphasis on energy or material recovery from waste, signifi-cantly complicating their overall environmental assessment. In orderto solve this problem, LCA methodology provides a powerful frame-work for the overall sustainability assessment of systems thatcombine different levels of energy and material recovery.

The comparative LCA of the two systems (results in Table 4)shows that, when separate collection is limited to 25%, the systemwith the incineration plant provides much better environmentalresults in all impact categories, especially if an electricity mix verydependent on fossil fuels is assumed for the background system. Butthe results change drastically if the comparison is performedconsidering that separate collection reaches 75%. This level ofseparate collection is supported by evidence in municipalities ofthe province of Gipuzkoa like Hernani, where the increase of separatecollection up to 80%, in conjunction with other waste preventionstrategies, has also carried with it important reductions in householdwaste generation. Under these conditions the system that empha-sizes separate collection and material recovery obtains better resultsin all impact categories but eutrophication, when compared to thesystemwith the WtE plant. The improvement is especially significantin the category of abiotic resource consumption (þ58%), and in thecategory of global warming potential (þ132% better).

The breakdown of each category result into partial contribu-tions from waste management stages and treatment processes

Fig. 4. Comparison of significant impact categories of scenarios A25 (energyrecovery from 75% mixed residual waste, material recovery from 25% separatelycollected waste) and B75 (material recovery from 75% separately collected waste,aerobic MBP of 25% mixed residual waste), broken down into partial contributionsin each category from waste management stages and treatment processes thatmake up both systems.

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459456

shows the importance of the avoided burdens in material recoveryfrom the separately collected plastics, paper, glass and metals.Under the conditions assumed in this work for the functional unitoperating in Gipuzkoa, it can be concluded that separately collect-ing a high share of waste—which thereby can be derived torecycling processes for material recovery—provides better envir-onmental results than deriving it as a mixed residue to anincineration plant where energy is recovered in the form ofelectricity. These superior environmental results are obtained evenin impact categories tightly related to fossil energy consumption,such as the global warming potential category. The only impactcategory in which the system with the incineration plant performsbetter is eutrophication, due to ammonia emissions in compostingof biowaste.

Besides, both systems generate similar final fluxes to landfill:7790 metric tons in the system with the incineration plant, versus9939 metric tons in the system without incineration. This showsthat spreading separate collection and promoting waste preven-tion may be such a good strategy as well as incinerating mixedresidual waste in order to reduce the quantity of residues finallyderived to landfill.

Acknowledgment

This research was supported by the Provincial Government ofGipuzkoa (R&D Research Contract 2012.0485, “Hiri hondakineiburuzko txostena, haien tratamendu eta kudeaketa GipuzkoakoLurrandean”).

Appendix A. System expansion to determine avoided burdens

System expansion/subtraction is performed to solve the alloca-tion of impacts and benefits of different systems that are intrinsi-cally multifunctional. It is performed as follows.

Fig. A.1(a) shows a diagram of waste-management system i forthe treatment of waste Wi (the Service that determines thefunctional unit); the system also produces a series of complemen-tary products (Pj,i), and causes some specific impacts. In our studywe perform a screening LCA in which we focus on abiotic resourcedepletion (ardi) and global warming potential (gwi) impact cate-gories, as they are considered to show the following trend of mostimportant environmental impact categories [56]. RMi is theresource material demand for the functioning of system i, andPEi corresponds to primary energy demand, which is analogous to

the Cumulative Energy Demand impact assessment methodimplemented in the Ecoinvent database [68].

Multifunctionality is solved by system expansion [69]. In a firststep, system expansion is performed in all compared systems untilall expanded systems produce identical quantities of commonproducts and services. Such system expansion is performed ineach system for each product Pj, making use of the correspondingproduction blocks for each product (Fig. A.1(b)), in which produc-tion inputs and corresponding impacts are recorded. In coherencewith the attributional modeling principle, average processes in thebackground system are considered for their characterization.Secondly, production outputs and inputs related to all coproductscomplementary to the main service provided by the waste-management system are subtracted from all expanded systems,using again the average processes in the background system. Thesetwo steps can be condensed in just one step in which productionof every complementary coproduct is subtracted in each systemusing the energy and material input demand and environmentalimpacts that correspond for the production of each complemen-tary product in the background system; the net result is shownschematically in Fig. A.1(c).

The multifunctionality problem is solved in attributional LCAby the accounting as avoided burdens of those impacts associatedwith the production in the background system, with some specificaverage processes, of the products substituted by the complemen-tary coproducts. This way, a correct characterization of theseaverage processes is critical; actually, these avoided burdens areso important that net environmental impacts are usually negativein most systems and for most indicators: the net environmentalbalance of the waste-management system results to be beneficialdue to the substitution of other more harmful ways to produce thecoproducts in the background system complementary to the wastemanagement service.

When systems expansion/substitution is performed in order tosolve the multifunctionality problem, with the crediting of avoidedburdens, it is not fair to compare different waste-managementsystems in terms of direct energy generation or direct materialrecovery. When different systems (Fig. A.1(a)) are credited withthe avoided burdens associated to the production of the copro-ducts in each system (Fig. A.1(b)), the resultant systems that weare actually comparing through the LCA neither produce energynor recover materials (Fig. A.1(c)), and consequently it is notadequate to compare those systems in terms of directly generatedelectricity, or of quantities of recycled materials. At best, a faircomparison of produced coproducts should be made through theexpanded systems; but the result is previously known: allexpanded systems under comparison must provide exactly thesame coproducts—altogether with the service of the functionalunit—, as that is actually the condition imposed to solve themultifunctionality problem, indispensable to allow a fair compar-ison of environmental impacts. A similar argument is applicablewhen we refer to efficiency, e.g. of electricity generation. Theefficiency of a waste-management system with an incinerationplant that presents a thermoelectric efficiency of 25% is not betterthan that of an expanded system that lacks incineration plants, asthe efficiency of the latter is precisely the one of the backgroundsystem, i.e. a power system with highly optimized units [43].

From the previous reasoning, however, we may not concludethat energy and material recovery is neither considered norquantified in comparative LCA. Indeed, they are accounted throughthe avoided burdens linked to the production of the materials andenergy substituted by the coproducts, and thus credited to thesystems. As shown in Fig. A.1(c), the substitution of material Pjwith a recycled material in system i is credited with a negativeimpact �ardPj,i in the field of abiotic resource depletion, and anegative impact �gwPj,i in the field of global warming, e.g. due to

Fig. A.1. (a) System i for treatment of waste Wi, which also produces a series ofcomplementary products (Pj,i), and causes some specific impacts ardi and gwi;(b) production system of product j to be considered in expanded systems, whichrequires of resource materials (RMPj) and primary energy (PEPj), and causes impacts(ardPj, gwPj); (c) waste-management system i in which complementary coproductsand corresponding inputs and impacts have been subtracted.

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459 457

the avoided consumption of fossil fuels needed to obtain productPj in the background system.

When these avoided burdens are credited, after subtraction,they also appear among the inputs to the compared systems.System i is credited with a negative input of resource materials(�RMPj,i) and primary energy (�PEPj,i) due to the avoided con-sumption of materials and energy otherwise required to obtain theproduct/material Pj, substituted by a particular recovered copro-duct. For the case of primary energy, the term PEi–ΣjPEPj,icorresponds to the net primary energy demand of system isubtracted the coproducts—which is analogous to applying theCumulative Energy Demand impact assessment method imple-mented in the Ecoinvent database [68]. RMi–ΣjRMPj,i correspondsto the net resource material demand for the functioning of systemi, subtracted the coproducts.

Net material and energy demands may be negative in thiscalculation, as they correspond to a subtracted system that iscredited with some avoided burdens, and those may be significant.This negative net input flux of energy and materials, however,should not be interpreted as a net positive output flux, as we areconsidering subtracted (differential) systems. Its effect in theoverall balance is normally reflected through the monetization[14] of energy and materials recovered by the waste-managementsystem, which, through their market values, internalize the pri-mary energy and resource materials required for their productionor fabrication in the background system [23,70].

References

[1] Diputación Foral de Gipuzkoa. AURRERAPEN DOKUMENTUAREN GARAPENER-AKO ESTRATEGIA 2008–2016 ESTRATEGIA DE DESARROLLO DEL DOCUMENTODE PROGRESO 2008–2016, ⟨http://www.gipuzkoaberri.net/WAS/CORP/DPDOficinaPrensaDigitalWEB/descarga.do?121110117116066073043104116115121106115110105116066070043104116105110108116066055043116119105106115066054⟩; 2012 [accessed 20.05.14].

[2] Diputación Foral de Bizkaia. Observatorio Permanente de Residuos Urbanosdel Territorio Histórico de Bizkaia. Datos de residuos, http://www.bizkaia.net/home2/Temas/DetalleTema.asp?Tem_Codigo¼7709&idioma¼CA&bnetmobile¼0&dpto_biz¼9&codpath_biz-¼9|351|7709; 2012, 2013, 2014 [accessed 20.05.14].

[3] Mijangos F. Urban-rural duality and waste management. In: Proceedings ofthe Klimagune workshop on opportunities and challenges for rural areas inthe context of climate change, ⟨http://www.bc3research.org/klimagune/images/stories/workshop/2014/ponencias/KW2014_Fernando_Mijangos.pdf⟩;2014 [accessed 17.03.15].

[4] Eurostat. Municipal waste Database, ⟨http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=env_wasmun&lang=en⟩ [accessed 17.03.15].

[5] Diputación Foral de Gipuzkoa. PIGRUG 2002-2016, Documento de Progreso2008–2016, ⟨http://www4.gipuzkoa.net/medioambiente/dpro/doc/es/01Documento_de_Progreso_CAST.pdf⟩; 2008 [accessed 20.05.14].

[6] Consorcio de Residuos de Gipuzkoa. Tablas de datos de los Residuos Urbanos deGipuzkoa, ⟨http://www.ghk.eus/es/datos/gipuzkoa/gipuzkoa-2013⟩ [accessed16.01.15].

[7] Muñoz I, Rieradevall J, Doménech X, Milà L. LCA application to integratedwaste management planning in Gipuzkoa (Spain). Int J Life Cycle Assess2004;9:272–80. http://dx.doi.org/10.1007/BF02978603.

[8] European Commission. Directive 2008/98/EC of the European Parliament andof the Council of 19 November 2008 on waste and repealing certain Directives.Off J Eur Union L; 2008. p. 312.

[9] European Integrated Pollution Prevention and Control Bureau (EIPPCB). Bestavailable techniques (BAT) reference document for Waste Incineration; 2006.

[10] ISO 14040:2006. Environmental management – life cycle assessment – principlesand framework. Brussels: CEN (European Committee for Standardisation).

[11] ISO 14044:2006. Environmental management – life cycle assessment – require-ments and guidelines. Brussels: CEN (European Committee for Standardisation).

[12] Guinée JB, Gorrée M, Heijungs R, Huppes G, Kleijn R, de Koning A. Handbookon life cycle assessment: operational guide to the ISO standards. Dordrecht,The Netherlands: Kluwer Academic Publisher; 2002.

[13] EC-JRC-IES. International reference life cycle data system (ILCD) handbook.General guide for life cycle assessment—detailed guidance. 1st ed. EuropeanCommission–Joint Research Centre–Institute for Environment and Sustain-ability; 2010.

[14] EC-JRC-IES. Supporting environmentally sound decisions for waste manage-ment—a technical guide to life cycle thinking (LCT) and life cycle assessment(LCA) for waste experts and LCA practitioners. 1st ed. European Commission–Joint Research Centre–Institute for Environment and Sustainability; 2011.

[15] Gentil EC, Damgaard A, Hauschild M, Finnveden G, Eriksson O, Thorneloe S, et al.Models for waste life cycle assessment: review of technical assumptions. WasteManag 2010;30:2636–48. http://dx.doi.org/10.1016/j.wasman.2010.06.004.

[16] De Feo G, Malvano C. The use of LCA in selecting the best MSWmanagement system.Waste Manag 2009;29:1901–15. http://dx.doi.org/10.1016/j.wasman.2008.12.021.

[17] Bovea MD, Ibáñez-Forés V, Gallardo A, Colomer-Mendoza FJ. Environmentalassessment of alternative municipal solid waste management strategies. ASpanish case study. Waste Manag 2010;30:2383–95. http://dx.doi.org/10.1016/j.wasman.2010.03.001.

[18] Pires A, Chang N, Martinho G. Reliability-based life cycle assessment for futuresolid waste management alternatives in Portugal. Int J Life Cycle Assess2011;16:316–37. http://dx.doi.org/10.1007/s11367-011-0269-7.

[19] Tunesi S. LCA of local strategies for energy recovery from waste in England,applied to a large municipal flow. Waste Manag 2011;31:561–71. http://dx.doi.org/10.1016/j.wasman.2010.08.023.

[20] Slagstad H, Brattebø H. LCA for household waste management when planninga new urban settlement. Waste Manag 2012;32:1482–90. http://dx.doi.org/10.1016/j.wasman.2012.03.018.

[21] Song Q, Wang Z, Li J. Environmental performance of municipal solid wastestrategies based on LCA method: a case study of Macau. J Clean Prod2013;57:92–100. http://dx.doi.org/10.1016/j.jclepro.2013.04.042.

[22] Bernstad A, la Cour Jansen J. Review of comparative LCAs of food wastemanagement systems – current status and potential improvements. WasteManag 2012;32:2439–55. http://dx.doi.org/10.1016/j.wasman.2012.07.023.

[23] Eriksson O, Carlsson Reich M, Frostell B, Björklund A, Assefa G, Sundqvist JO,et al. Municipal solid waste management from a systems perspective. J CleanProd 2005;13:241–52. http://dx.doi.org/10.1016/j.jclepro.2004.02.018.

[24] Merrild H, Larsen AW, Christensen TH. Assessing recycling versus incinerationof key materials in municipal waste: the importance of efficient energyrecovery and transport distances. Waste Manag 2012;32:1009–18. http://dx.doi.org/10.1016/j.wasman.2011.12.025.

[25] Nadzirah Othman S, Zainon Noor Z, Halilu Abba A, Yusuf RO, Abu Hassan MAriffin. Review on life cycle assessment of integrated solid waste managementin some Asian countries. J Clean Prod 2013;41:251–62. http://dx.doi.org/10.1016/j.jclepro.2012.09.043.

[26] Gentil EC, Gallo D, Christensen TH. Environmental evaluation of municipalwaste prevention. Waste Manag 2011;31:2371–9. http://dx.doi.org/10.1016/j.wasman.2011.07.030.

[27] Bernstad A, la Cour Jansen J. A life cycle approach to the management ofhousehold food waste – a Swedish full-scale case study. Waste Manag2011;32:1879–96. http://dx.doi.org/10.1016/j.wasman.2011.02.026.

[28] Koroneos CJ, Nanaki EA. Integrated solid waste management and energy produc-tion—a life cycle assessment approach: the case study of the city of Thessaloniki. JClean Prod 2012;27:141–50. http://dx.doi.org/10.1016/j.jclepro.2012.01.010.

[29] Buttol P, Masoni P, Bonoli A, Goldoni S, Belladonna V, Cavazzuti C. LCA ofintegrated MSW management systems: case study of the Bologna District. WasteManag 2007;27:1059–70. http://dx.doi.org/10.1016/j.wasman.2007.02.010.

[30] Rigamonti L, Grosso M, Giugliano M. Life cycle assessment for optimising thelevel of separated collection in integrated MSW management systems. WasteManag 2009;29:934–44. http://dx.doi.org/10.1016/j.wasman.2008.06.005.

[31] Calabrò PS. Greenhouse gases emission from municipal waste management:the role of separate collection. Waste Manag 2009;29:2178–87. http://dx.doi.org/10.1016/j.wasman.2009.02.011.

[32] Consonni S, Giugliano M, Massarutto A, Ragazzi M, Saccani C. Material andenergy recovery in integrated waste management systems: project overviewand main results. Waste Manag 2011;31:2057–65. http://dx.doi.org/10.1016/j.wasman.2011.04.016.

[33] Cimpan C, Wenzel H. Energy implications of mechanical and mechanical–biological treatment compared to direct waste-to-energy. Waste Manag2013;33:1648–58. http://dx.doi.org/10.1016/j.wasman.2013.03.026.

[34] Belboom S, Digneffe JM, Renzoni R, Germain A, Léonard A. Comparingtechnologies for municipal solid waste management using life cycle assess-ment methodology: a Belgian case study. Int J Life Cycle Assess2013;18:1513–23. http://dx.doi.org/10.1007/s11367-013-0603-3.

[35] Koci V, Trecakova T. Mixed municipal waste management in the CzechRepublic from the point of view of the LCA method. Int J Life Cycle Assess2011;16:113–24. http://dx.doi.org/10.1007/s11367-011-0251-4.

[36] Scipioni A, Mazzi A, Niero M, Boatto T. LCA to choose among alternative designsolutions: the case study of a new Italian incineration line. Waste Manag2009;29:2462–74. http://dx.doi.org/10.1016/j.wasman.2009.04.007.

[37] Wittmaier M, Langer S, Sawilla B. Possibilities and limitations of life cycleassessment (LCA) in the development of waste utilization systems – appliedexamples for a region in Northern Germany. Waste Manag 2009;29:1732–8.http://dx.doi.org/10.1016/j.wasman.2008.11.004.

[38] Assamoi B, Lawryshyn Y. The environmental comparison of landfilling vs.incineration of MSW accounting for waste diversion. Waste Manag2012;32:1019–30. http://dx.doi.org/10.1016/j.wasman.2011.10.023.

[39] Burnley S, Phillips R, Coleman T, Rampling T. Energy implications of thethermal recovery of biodegradable municipal waste materials in the UnitedKingdom. Waste Manag 2011;31:1949–59. http://dx.doi.org/10.1016/j.wasman.2011.04.015.

[40] Fruergaard T, Astrup T. Optimal utilization of waste-to-energy in an LCAperspective. Waste Manag 2011;31:572–82. http://dx.doi.org/10.1016/j.wasman.2010.09.009.

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459458

[41] den Boer J, den Boer E, Jager J. LCA-IWM: a decision support tool forsustainability assessment of waste management systems. Waste Manag2007;27:1032–45. http://dx.doi.org/10.1016/j.wasman.2007.02.022.

[42] ALTAIR Ingeniería. Documento 1/2: Metodología y caracterización de lafracción resto de los residuos domésticos generados en hogares y comercios,y la fracción resto de los residuos comerciales que se depositan en el mismocontenedor, para el Territorio Histórico de Gipuzkoa, ⟨http://www4.gipuzkoa.net/MedioAmbiente/gipuzkoaingurumena/adj/documentacion/CARACTERIZACION%202012-2013.pdf⟩; 2013 [accessed 20.05.14].

[43] Consonni S, Viganò F. Material and energy recovery in integrated wastemanagement systems: the potential for energy recovery. Waste Manag2011;31:2074–84. http://dx.doi.org/10.1016/j.wasman.2011.05.013.

[44] Household collection schedule in Bilbao, ⟨http://www.bilbao.net/cs/Satellite?c=BIO_Servicio_FA&cid=3007556277&language=es&pageid=3000094417&pagename=Bilbaonet%2FBIO_Servicio_FA%2FBIO_Servicio&anclaServ=aB3&rutaCatServ=3003446956⟩ [accessed 20.05.14].

[45] Household collection schedule in Donosita-San Sebastián, ⟨http://jokogarbia.donostia.org/es/errefusa/⟩ [accessed 20.05.14].

[46] Ayuntamiento de Hernani. Datos de recogida 2010, ⟨http://www.hernani.net/images/stories/zerbitzuak/Atez_ate/2010eko_datuak.pdf⟩ [accessed 20.05.14].

[47] Ayuntamiento de Hernani. Datos de recogida 2011, ⟨http://www.hernani.net/images/stories/zerbitzuak/Atez_ate/2011KO_BILKETAREN__DATUAK.pdf⟩[accessed 20.05.14].

[48] Mancomunidad de SanMarkos. Datos-oficiales-municipios09 (Hernani).xls,personal communication; 2013.

[49] Ekvall T, Assefa G, Björklund A, Eriksson O, Finnveden G. What life-cycleassessment does and does not do in assessments of waste management. WasteManag 2007;27:989–96. http://dx.doi.org/10.1016/j.wasman.2007.02.015.

[50] Cleary J. The incorporation of waste prevention activities into life cycleassessments of municipal solid waste management systems: methodologicalissues. Int J Life Cycle Assess 2010;15:579–89. http://dx.doi.org/10.1007/s11367-010-0186-1.

[51] Nessi S, Rigamonti L, Grosso M. Discussion on methods to include preventionactivities in waste management LCA. Int J Life Cycle Assess 2013;18:1358–73.http://dx.doi.org/10.1007/s11367-013-0570-8.

[52] Hong J, Li X, Zhaojie C. Life cycle assessment of four municipal solid wastemanagement scenarios in China. Waste Manag 2010;30:2362–9. http://dx.doi.org/10.1016/j.wasman.2010.03.038.

[53] Papageorgiou A, Barton JR, Karagiannidis A. Assessment of the greenhouseeffect impact of technologies used for energy recovery from municipal waste:a case for England. J Environ Manag 2009;90:2999–3012. http://dx.doi.org/10.1016/j.jenvman.2009.04.012.

[54] Ekvall T, Tillman A, Molander S. Normative ethics and methodology for lifecycle assessment. J Clean Prod 2005;13:1225–34. http://dx.doi.org/10.1016/j.jclepro.2005.05.010.

[55] European Commission. Communication from the Commission to the Europeanparliament, the Council, the European Economic and social committee and theCommittee of the regions: a roadmap for moving to a competitive low-carboneconomy in 2050, European commission SEC; 2011 288 final.

[56] CO2Scorecard, ⟨http://www.co2scorecard.org/⟩; 2014 [accessed 20.05.14].

[57] Red Eléctrica de España. El sistema eléctrico español, AVANCE DEL INFORME2013, ⟨http://www.ree.es/sites/default/files/downloadable/avance_informe_sistema_electrico_2013.pdf⟩; 2013 [accessed 20.05.14].

[58] Swiss Centre for Life Cycle Inventories. ECOINVENT-2000 Data V1.01, LCI ofelectricity supply mix in European countries, ⟨http://www.ecoinvent.org/database/⟩; 2003 [accessed 20.05.14].

[59] den Boer E, den Boer J, Jager J, Rodrigo J, Meneses M, Castells F et al.Deliverable report on D3.1 and D3.2: environmental sustainability criteriaand indicators for waste management (Work Package 3). The use of life cycleassessment tool for the development of integrated waste managementstrategies for cities and regions with rapid growing economies LCA-IWM;2005.

[60] Reimann DO. CEWEP energy report III (Status 2007–2010) results of specificdata for energy, R1 plant efficiency factor and NCV of 314 European waste-to-energy (WtE) plants. Bamberg, Germany, ⟨http://www.cewep.eu/m_1069⟩;2012 [accessed 20.05.14].

[61] Gobierno Vasco. Proyecto técnico y estudio de impacto ambiental del Centrode Gestión de Residuos de Gipuzkoa, ⟨http://www.ingurumena.ejgv.euskadi.net/r49-6172/es/contenidos/informe_estudio/gipuzkoako_hondakin_kudeaketa/es_doc/inicio.html⟩; 2009 [accessed 20.05.14].

[62] Arafat HA, Jijakli K, Ahsan A. Environmental performance and energy recoverypotential of five processes for municipal solid waste treatment. J Clean Prod2013 http://dx.doi.org/10.1016/j.jclepro.2013.11.071, in press.

[63] Fricke K, Bahr T, Bidlingmaier W, Springer C. Energy efficiency of substanceand energy recovery of selected waste fractions. Waste Manag 2011;31:644–8.http://dx.doi.org/10.1016/j.wasman.2010.11.017.

[64] Saer A, Lansing S, Davitt NH, Graves RE. Life cycle assessment of a food wastecomposting system: environmental impact hotspots. J Clean Prod2013;52:234–44. http://dx.doi.org/10.1016/j.jclepro.2013.03.022.

[65] Cadena E, Colón J, Artola A, Sánchez A, Font X. Environmental impact of twoaerobic composting technologies using life cycle assessment. Int J Life CycleAssess 2009;14:401–10. http://dx.doi.org/10.1007/s11367-009-0107-3.

[66] Eurostat, 2012. Guidance on municipal waste data collection, November 2012,Eurostat – Unit E3 – Environment and forestry, guidance on municipal wastedata collection November–2012, WASTE WG 5.2 b(2012), ⟨http://epp.eurostat.ec.europa.eu/portal/page/portal/waste/documents/Municipal_waste_statistics_guidance.pdf⟩; 2012 [accessed 20.05.14].

[67] Khoo HH, Tan LLZ, Tan RBH. Projecting the environmental profile of Singa-pore's landfill activities: comparisons of present and future scenarios based onLCA 2012;32:890–900. http://dx.doi.org/10.1016/j.wasman.2011.12.010WasteManag 2012;32:890–900. http://dx.doi.org/10.1016/j.wasman.2011.12.010.

[68] Frischknecht R, Jungbluth N, Althaus H, Bauer C, Doka G, Dones R, et al.Ecoinvent report no. 3. Implementation of life cycle impact assessmentmethods: data v2.0. Dübendorf, Switzerland: Swiss centre for Life CycleInventories; 2007.

[69] Weidema B. Avoiding co-product allocation in life-cycle assessment. J Ind Ecol2001;11:4–33.

[70] Massarutto A, de Carli A, Graffi M. Material and energy recovery in integratedwaste management systems: a life-cycle costing approach. Waste Manag2011;31:2102–11. http://dx.doi.org/10.1016/j.wasman.2011.05.017.

G. Bueno et al. / Renewable and Sustainable Energy Reviews 51 (2015) 449–459 459