P

LE

a

ARRAA

KCPCFB

1

(mJfa(tigvtthdetfdnuea2

1d

International Journal of Greenhouse Gas Control 6 (2012) 77–83

Contents lists available at SciVerse ScienceDirect

International Journal of Greenhouse Gas Control

j ourna l ho mepage: www.elsev ier .com/ locate / i jggc

ublic acceptance of CCS system elements: A conjoint measurement

asse Wallquist ∗, Selma L’Orange Seigo, Vivianne H.M. Visschers, Michael SiegristTH Zurich, Institute for Environmental Decisions (IED), Consumer Behavior, Universitätstrasse 22, 8092 Zurich, Switzerland

r t i c l e i n f o

rticle history:eceived 14 July 2011eceived in revised form 2 November 2011ccepted 11 November 2011vailable online 30 December 2011

a b s t r a c t

The aim of the present study is to examine public preferences regarding the characteristics of the threeelements of carbon dioxide capture and storage (CCS): capture, pipeline, and storage. A random sample of139 Swiss citizens received basic information about CCS online and then participated in an experiment.A conjoint measurement of CCS acceptance and analysis of variance was used to examine respondents’preferences for characteristics of CCS elements. This approach allowed respondents to make trade-offs

eywords:CSublic acceptanceonjoint analysisield trialsECCS

by expressing preferences for complete CCS systems instead of evaluating single elements in isolation.Our results show that people put most emphasis on pipelines near their homes and on the type of plantthe CO2 originates from. A “Not in my backyard (NIMBY) effect” was found both for pipelines and storage.This NIMBY effect, however, disappears when CO2 from a biogas-fired plant is used for the injection. Weconclude that it may be possible to avoid the NIMBY effect for geological storage field trials by usingbioenergy with carbon dioxide capture and storage (BECCS).

. Introduction

The way the public perceives carbon dioxide capture and storageCCS) is crucial to the technology’s further development and imple-

entation. Recently, a number of projects (e.g. Barendrecht (NL),änschwalde (DE)) aimed at mitigating CO2 emissions from fossiluel power plants by applying carbon dioxide capture and stor-ge (CCS) have been delayed or stopped due to public oppositionDesbarats et al., 2010; Verhagen, 2010). This indicated the impor-ance of understanding how people perceive CCS. Public oppositionnfluenced CCS projects either directly in the form of local actionroups, or indirectly by making the political climate for CCS unfa-orable. One problem that social science researchers face whenhey examine public perception of CCS is that in most countries,he public is rather unfamiliar with it. Only 10% of Europeans haveeard of CCS (European Commission, 2011). In Japan, 19% of respon-ents reported hearing about CCS or some knowledge of it (Itaokat al., 2009). Results from Miller et al. (2008) show that not morehan 18% of Australians have heard of geological storage of CO2 androm the United States, Curry et al. (2007) report that 5% of respon-ents have heard of CCS. These results show that most people doot have an opinion about CCS. Therefore, researchers need to sim-late the future by providing respondents with information before

liciting their attitudes. The stability of these attitudes that are cre-ted on the spot has been questioned (de Best-Waldhober et al.,009; Malone et al., 2010).

∗ Corresponding author. Tel.: +41 44 632 3207; fax: +41 44 632 1029.E-mail address: lwallquist@ethz.ch (L. Wallquist).

750-5836/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.oi:10.1016/j.ijggc.2011.11.008

© 2011 Elsevier Ltd. All rights reserved.

CCS is a complex technology. The public may therefore per-ceive many elements as hazardous (e.g., transport and storage). Allelements should, thus, be addressed in communication with stake-holder groups, public and private decision-makers, and laypeople.It is important to understand which elements influence percep-tions of CCS most, and whether the characteristics of one elementinfluence the perception of another one (e.g. does the perceptionof the storage element depend on the type of plant where CO2 iscaptured?) This has not been investigated in detail, but is relevantfor communicators and policy-makers, especially when there aretime and budget restraints to evaluate all CCS technology optionsin detail. Thus, investigating the contribution of these elements toacceptance in more detail is worthwhile. The aim of the presentstudy is, therefore, to study the importance of the three CCS ele-ments – capture, transport, and storage – for public acceptanceusing a conjoint measurement among laypeople in Switzerland.The conjoint measurement of acceptance allows respondents to dotrade-offs between CCS elements as participants compare completesystems instead of single elements in isolation.

A considerable amount of research has revealed factors thatinfluence risk perception and acceptance of CCS (e.g., Shackleyet al., 2005; Terwel et al., 2011; Wallquist et al., 2010). In theNetherlands, providing information that was accurate and under-standable for Dutch laypeople produced stable attitudes andshowed that informed laypeople accept CCS (de Best-Waldhoberet al., 2009). In the United States, researchers showed that informed

laypeople are willing to accept CCS as long as it is part of a portfolioof low-carbon energy technologies (Fleishman et al., 2010).

In terms of acceptance of the individual steps of CCS, researchersfound that for the capture part, people are mainly concerned about

7 l of G

tctWfGu(ewnrtta

rsCrapdwbtnppcp(t(ibsh2sdpbstcrctlepoc“eCqrn

tnbnt

8 L. Wallquist et al. / International Journa

he burning of fossil fuels and the associated emissions. Socioe-onomic concerns here play a larger role than concerns about theechnical aspects of CCS (Palmgren et al., 2004; Shackley et al., 2005;

allquist et al., 2009, 2010). In Switzerland, there are currently noossil-fueled power plants that feed electricity into the public grid.as-fired power plants, however, have been proposed as a back-p for renewable energy technologies after the nuclear phase-outBundesrat, 2011). There is considerable potential for sustainablenergy production from biomass (Steubing et al., 2010). Bio-energyith carbon dioxide capture and storage (BECCS) would result inegative CO2 emissions (Read and Lermit, 2005). This might be aelevant factor in shaping public opinion of CCS. Furthermore, theerm “bio,” which is used in German as equivalent to the Englisherm “organic,” may have a positive connotation for some peoplend result in a beneficial “halo-effect” (Thorndike, 1920).

To the best of our knowledge, the transport part of CCS haseceived no attention in experimental public perception researcho far, despite the fact that humans have some experiences withO2 pipelines. In the United States, CO2 pipeline systems have beenegulated and operated for a long time (Dooley et al., 2009). From

technical point of view, the transport of CO2 in pipelines is com-arable to that of natural gas, although some differences do exist,ue to different physical properties of the two gases. It is unclearhether laypeople distinguish between CO2 and gas pipelines,

ecause there is still little experience with public reactions to theransport of CO2 through populated areas, whereas transport ofatural gas is a common practice. If there were plans to trans-ort CO2 over long distances and through populated areas, someeople might be opposed to such pipelines in their vicinity if theyould not benefit from, for example, job opportunities at a nearbyower plant or financial compensation. There might be a NIMBY“not in my backyard”) effect. The NIMBY metaphor has been usedo explain the tension between general support for a technologye.g. wind farms) and local public opposition against particularnstallations of this technology. This may be the case where localenefits and local risk exposure are not equally distributed. Theimplified explanation of local opposition by mere NIMBYism has,owever, experienced considerable criticism (e.g., Devine-Wright,005; Wolsink, 2006). Wolsink (2000) showed that other factors,uch as general attitudes towards a technology, are stronger pre-ictors of willingness to oppose than simple NIMBYism. Publicerception of the storage part of CCS has been addressed by a num-er of studies on acceptance of CCS. A recent Eurobarometer studyhowed that the public is generally concerned about storage loca-ions within 5 km of their home (European Commission, 2011). Thisan be interpreted in terms of the NIMBY phenomenon. Among theeasons for laypeople’s rejection of CO2 storage in their vicinity areoncerns such as leakage, over-pressurization of the reservoir, andhe fear of negative effects on the natural environment and pub-ic health (Ashworth et al., 2010; Palmgren et al., 2004; Shackleyt al., 2005; Wallquist et al., 2010). Singleton et al. (2009) com-ared influential risk characteristics of geological CO2 storage withther hazards such as radioactive waste or fossil fuels. In the psy-hometric framework, these risk characteristics are grouped into anunknown” and a “dread” factor (Fischhoff et al., 1978). Singletont al.’s (2009) results suggest that risks from geological storage ofO2 would be perceived as not very high for the “dread” factor, butuite high for the “unknown” factor. This implies that the perceivedisk could be lowered by additional field trials, thus decreasing theewness of the risk for the public.

In areas where the public has only little or no experience withhe extraction of fossil fuels (e.g., Switzerland), the role of the

ewness of geological storage of CO2 for risk perception maye even more pronounced. After the accident at the Fukushimauclear power plant in 2011, political pressure on governmentso phase out nuclear energy production has increased in some

reenhouse Gas Control 6 (2012) 77–83

countries, including Switzerland (Bundesrat, 2011). Under suchsocial pressure, fossil-fueled power plants with CCS may contributeto sustaining a low-carbon energy supply, especially in countriesthat have no history of fossil-fueled electricity production. Earlylearning from field tests is, in this case, of particular importancefor science, but also for the public. A roadmap for testing CCS inSwitzerland, including a geological field test, has been introduced(Sutter et al., 2011). For science, field tests are important, becauseonly little previous geological data on potential reservoirs is avail-able. For the public, field tests are also important, because the publicis not used to technologies involving geological engineering andfossil-fueled power plants. Increased awareness and experienceswith the technology through new field trials may increase pub-lic support (Singleton et al., 2009). Before conducting field trialsof complete CCS systems, it makes sense to examine what ele-ments of CCS systems are important to the public. Thus far, onlylimited research about the role of the three CCS elements for publicacceptance of complete systems has been conducted. To the bestof our knowledge, the effects of interactions of these elements onacceptance have not been studied.

The current study aims at examining public preferences for thecharacteristics of the elements capture, transport, and storage incombination, and shedding some light on the relative importanceof each. To this end, we used a conjoint measurement. Conjointanalyses have been proposed for analyzing preferences for theoret-ical or factual products and systems. Participants assess a numberof complete product descriptions that differ with respect to mul-tiple factors. Conjoint analysis allows researchers to quantify therelative importance of different product factors, as well as the per-ceived utilities of different levels of these factors. The strength ofsuch a measurement is that it is similar to real-world choice situ-ations (Hair, 1995). It builds on the premise that respondents canbetter express their preferences for a complete product or a systemthan for separated single factors. Thereby, the factors and their cor-responding levels should be communicable and technically feasible(Hair, 1995). The choice of factors should furthermore be theoreti-cally motivated. In marketing research, conjoint analysis has foundwidespread application (Green and Srinivasan, 1990). It has alsobeen used for assessing public acceptance of genetic engineeringin food processing and of nanotechnology in food and packagingas well as for examining public preferences for base station sitingfor mobile communication (Dohle et al., 2010; Frewer et al., 1997;Siegrist et al., 2009). Alriksson and Öberg (2008) concluded thatconjoint analysis is a useful method in environmental risk anal-ysis and communication because these fields would benefit fromthe same strengths of conjoint analysis as traditional marketingresearch. Trade-offs between different elements can be examinedby evaluating and comparing complete solutions.

2. Method

2.1. Sample

We conducted an online experiment with participants fromthe general public in the German-speaking part of Switzerland inspring 2011. Respondents were recruited by from an online panel,consisting of persons who had previously volunteered to regu-larly participate in different kinds of scientific studies conductedby ETH Zurich. The invitation for the present experiment did notmention the exact topic of the study, in order to avoid people drop-ping out because they thought they did not know enough about

the technologies. After excluding 34 respondents who answeredall questions with the same score (i.e., had zero variance in theirresponses) or who filled in their responses in an unrealisticallyshort amount of time (c.f. Carver, 1990), we included 139 people

L. Wallquist et al. / International Journal of Greenhouse Gas Control 6 (2012) 77–83 79

Table 1Factors and corresponding levels used to combine hypothetical living situations.

Level 1 Level 2 Level 3

Type of plant Biogas-fired Gas-fired –Storage location In own municipality In a neighboring –

iwt5t

2

ntaptTaf

mocg

dmdtta

rfssco

2

iodfftWlt

fiptfte

Table 2Means and standard deviations of the acceptance scores for the different scenarios,in descending order.

Scenario (type of power plant where CO2 iscaptured/storage location/type of pipeline)

M SD

Biogas-fired, in neighboring canton, no pipeline 72 29Biogas-fired, in own municipality, no pipeline 67 30Gas-fired, neighboring canton, no pipeline 66 31Biogas-fired, neighboring canton, CO2-pipeline 57 30Gas-fired, in own municipality, no pipeline 56 32Biogas-fired, in own municipality, CO2-pipeline 55 31Biogas-fired, in own municipality, gas-pipeline 52 30Biogas-fired, in neighboring canton, gas-pipeline 50 31Gas-fired, in neighboring canton, CO2-pipeline 49 30Gas-fired, in own municipality, CO2-Pipeline 47 31Gas-fired, in neighboring canton, gas-pipeline 45 32Gas-fired, in own municipality, gas-pipeline 43 31

acceptance preferences emerges.

cantonType of pipeline Gas pipeline CO2 pipeline No pipeline

n our analysis. Thirty-seven percent (n = 51) of the respondentsere female. Six persons included in the analysis did not report

heir gender. Age ranged from 19 to 87 years. The mean age was4.82 years (SD = 13.87). Thirty-two percent (n = 45) reported thathey had a college or university degree.

.2. Information section

Most people are not familiar with CCS. Our web-based question-aire, therefore, started with a short introduction to the topic sohat participants would have a basic understanding of what CCS isbout. The presented section is based on a text that had been used inrevious studies (L’Orange Seigo et al., 2011; Wallquist et al., 2010)o inform participants about the principal reasoning behind CCS.he section was carefully worded, reviewed by experts, and aimedt briefing respondents in plain language about the basic aim andunctioning of CCS technologies. Specific risks were not discussed.

The introductory text read as follows (translated from German):The increase of carbon dioxide (CO2) in the atmosphere is the

ost important factor in climate change. Every day, large amountsf CO2 are generated. This is happening, for example, through theombustion of gas in power plants. Such plants are supplied withas by subterranean pipelines.

To mitigate climate change, CO2 emissions need to be reducedrastically. Carbon dioxide capture and storage is a technologicaleasure for reducing CO2 in the atmosphere. With this proce-

ure, CO2 is captured from the exhaust gases of power stations,ransported by subterranean pipelines to suitable storage loca-ions, and stored underground. Thereby, it can be kept out of thetmosphere.

Permanent storage of CO2 is possible only in certain porousock formations that are more than 800 m under the Earth’s sur-ace. Examples of such formations are depleted oil or gas fields andaline rock formations. The CO2 is liquefied and piped into thesetorage locations. From Swiss plants, no CO2 is stored undergroundurrently. In some European countries, smaller facilities alreadyperate. The technology is still in the development stage.

.3. Acceptance measurement

After reading the introduction text, respondents receivednstructions on how to fill in the questionnaire. They were told thatn the next page, they would be asked to assess descriptions of 12ifferent living situations. All living situations had in common theact that in their closer or further surroundings, there were plantsor the capture and storage of CO2, but they differed in terms ofhree factors: type of plant, storage location, and type of pipeline.

e presented the three factors and the corresponding levels in aist to all participants. Table 1 gives an overview of the factors andhe corresponding levels.

For the factor Plant, we chose the levels gas-fired and biogas-red, as these would be feasible types of carbonaceous fuel powerlants in Switzerland. For the factor Storage Location, we varied

he distance of the injection from the domicile in order to accountor the NIMBY effect. The NIMBY effect for pipelines and poten-ial differences in the perception of CO2 and of natural gas werexamined with the levels chosen for the factor Pipeline (Table 1).

The combination of the three factors and their levels resulted in 12scenarios. To be able to analyze interaction effects, we chose a fulldesign (i.e., assessment of acceptance for all combinations of factorlevels). This, however, limited the number of levels that could beexamined.

We then presented the following exemplary living situation1:“CO2 from a biogas-fired power plant is stored in the subsurface

of your municipality. Near your house passes a CO2 pipeline that isconnected to the power station.”

In the four scenarios without a pipeline, its absence wasexplicitly mentioned. An example description reads: “CO2 froma biogas-fired power plant is stored in the subsurface of yourmunicipality. Near your house passes no pipeline.”

Respondents were asked to rate the acceptability of each of the12 hypothetical living scenarios, combined from the levels shownin Table 1 and presented in the same way as shown in the exam-ple above. The sequence of the presented living scenarios wasrandomized for every respondent to avoid order effects. We mea-sured levels of acceptance with sliders with 101 points. These gaverespondents the impression that they could express their accep-tance levels on a continuum. Thereby, we avoided ceiling effects.The endpoints of these scales were labeled 1 = not at all acceptableand 101 = totally acceptable. At the end of the questionnaire, wecollected demographic information and thanked the participants.

3. Results

3.1. Acceptance scores

Each respondent had 12 acceptance scores, one for each sce-nario. Across participants, the full range of possible answers wasused, with acceptance scores ranging from 1 to 101 in each sce-nario. The mean values of each scenario ranged from 43 to 72.Table 2 shows the means and standard deviations of the acceptancescores for each scenario. The mean scores of the scenarios show thatscenarios that do not involve a pipeline near participants’ homes,and are generally more accepted than the ones involving CO2 orgas pipelines. For plants, scenarios with biogas-fueled plants arehigher ranked than those with conventional gas (named just “gas”in the study). For different storage locations, no clear picture of

1 The words underlined are the factor levels that differed between the hypothet-ical living scenarios (see Table 1 for factor levels).

80 L. Wallquist et al. / International Journal of Greenhouse Gas Control 6 (2012) 77–83

3

AfnFPfSstPogp

iPw(ia

iis

Fig. 1. Mean acceptance levels for different plants and storage locations.

.2. Analysis of variance

A 2 × 2 × 3 (Plant × Storage × Pipeline) repeated-measuresNOVA was computed in order to assess the effect of each

actor on respondents’ acceptance judgments individually. Sig-ificant main effects were observed for all three factors: Plant,(1, 138) = 23.77, p < 0.001; Storage, F(1, 138) = 7.89, p < 0.01; andipeline, F(1, 138) = 39.34, p < 0.001. The largest effect size wasound for Pipeline (�2 = 0.12), followed by Plant (�2 = 0.03) andtorage (�2 = 0.005). Apparently, people’s judgments were mosttrongly influenced by the factor Pipeline, to a lesser degree byhe factor Plant, and least of all by the factor Storage Location.airwise comparisons showed that acceptance of scenarios with-ut a pipeline were higher than the acceptance of scenarios withas (p < 0.001) or CO2 (p < 0.001) pipelines. The acceptance of CO2ipelines was higher than the acceptance of gas pipelines (p < 0.01).

Significant results were also found for two of the two-waynteraction terms: Plant × Storage, F(1, 138) = 7.98, p < .01, andipeline × Storage, F(1, 138) = 10.72, p < 0.001. Effect sizes, however,ere rather small, with �2 = 0.001 (Plant × Storage) and �2 = 0.005

Pipeline × Storage), respectively. Figs. 1 and 2 illustrate these twonteractions. The three-way interaction between Pipeline, Plant,nd Storage was not significant, F(1, 138) = 3.08, p = 0.082.

As can be seen in Fig. 1, storage of CO in the neighboring canton

2s generally more accepted than in the respondent’s own munic-pality. The difference is only pronounced, however, if the CO2ource is a (conventional) gas-fired power plant. If the power plant

Fig. 2. Mean acceptance levels for different pipelines and storage locations.

Fig. 3. Utilities for all factor levels.

is biogas-fired, acceptance levels converge, and participants findCO2 storage in their own municipality nearly as acceptable as in aneighboring canton.

Almost the opposite effect can be observed for the interac-tion of Storage with Pipeline (Fig. 2). Here, acceptance levels aresimilar, whether CO2 is stored in people’s own municipality ora neighboring canton, as long as there is a pipeline that carriesCO2 or natural gas near respondents’ homes. Assessments diverge,however, when no pipeline passes near people’s homes. In the lat-ter case, acceptance increases for both storage locations, but thisincrease is stronger in case of having the storage location in a neigh-boring canton than in the respondent’s own municipality.

3.3. Utilities for factor levels and relative importance of factors

A conjoint analysis was conducted, based on the participants’acceptance scores for the 12 living scenarios, to assess the impor-tance of the factors and the utilities of the seven levels. Utilitiesfor the different factor levels were calculated with SPSS CONJOINT.These utility values are similar to regression coefficients and offera quantification of the preferences for the factor levels, with highervalues indicating greater preferences (SPSS, 2007).

Fig. 3 gives an overview of the calculated utilities of all factorlevels. The values show that participants have higher utilities fora storage location in a neighboring canton compared to storagein their own municipality. Results also show a positive utility forbiogas-fired plants in comparison with gas-fired plants. Concern-ing the factor pipeline, both gas and CO2 pipelines have negativeutilities, whereas the utility of the level with no pipeline is positive.The utility for a CO2 pipeline is, however, not as negative as the onefor a gas pipeline.

Dividing the range of utilities for each separate factor by the sumof the utility ranges of all factors results in importance values for thefactors. These give an indication of how important the factor is forthe preference of the living scenario (SPSS, 2007). The factor withthe highest importance was Pipeline (62%) followed by Plant (27%)and Storage Location (11%). This means that the factor Pipeline wasalmost six times more important than the factor Storage Locationfor participants’ preferences for the living scenarios.

3.4. Cluster analysis

The utility scores yielded by the conjoint analysis were used to

perform a cluster analysis in order to identify subgroups of respon-dents. Following the recommendations of Backhaus et al. (2006) (p.583) for clustering utility values, a hierarchical, complete-linkagecluster analysis was conducted, using Pearson’s r as a proximity

L. Wallquist et al. / International Journal of G

Table 3Mean utility scores for each factor level in the different clusters.

Cluster 1: focuson pipelines(n = 41)

Cluster 2: focuson plant type(n = 43)

Cluster 3: focuson storage(n = 55)

Storage locationIn own municipality −3.1 2.6 −3.6In neighboring canton 3.1 −2.6 3.6

Type of pipelineCO2 pipeline −6.6 −2.1 −0.8Gas pipeline −21.3 −5.0 0.7No pipeline 27.9 7.0 0.1

mtstvtintpfs“tfs

4

tEozcao(wtsat

e–aop

jfitams2mllp

Type of plantBiogas-fired 1.0 10.7 0.6Gas-fired −1.0 −10.7 −0.6

easure. This yielded three distinct and fairly equally sized clus-ers. Each cluster is marked by particularly high (or low) utilitycores for one of the three factors (storage location, type of pipeline,ype of plant), as shown in Table 3. Participants in cluster 1 have aery high utility score for “no pipeline”. They are especially averseo gas pipelines, but considerably less so to CO2 pipelines. Personsn cluster 1 also exhibit a certain preference for CO2 storage in aeighboring canton as opposed to their own municipality. Clus-er 2 shows a strong preference for CO2 from biogas-fueled powerlants as opposed to CO2 from conventional gas. Its members pre-er “no pipeline” scenarios, while not showing extreme aversion tocenarios that do involve pipelines. People in this cluster show noNIMBYism” and even have slightly positive utilities for storage inheir own town. The biggest group, cluster 3, seems largely indif-erent towards pipelines and the origin of the CO2, but expresses amall preference for CO2 to be stored in a neighboring canton.

. Discussion

Deployments of carbon dioxide capture and storage (CCS) sys-ems are threatened considerably by public protests. Especially inurope, some planned CCS projects and even efforts for explorationf storage potentials have evoked substantial protest among citi-ens (Desbarats et al., 2010; Dütschke, 2011). Previous researchompared the risk characteristics of CCS with other potential haz-rds and showed that the newness of CCS is likely to act negativelyn the public’s risk perception and called for more field trialsSingleton et al., 2009). When designing new field trials, it may beorthwhile to consider if it is technically and economically feasible

o take the public’s preferences for characteristics of different CCSystems into account. We therefore used an analysis of variancend a conjoint analysis of different CCS system layouts to examinehe importance of the characteristics of CCS systems.

Our results show that for the studied scenarios with the threelements – type of plant, type of pipeline, and storage location

pipeline has the highest importance for laypeople, and stor-ge location the lowest. For pipelines, a NIMBY effect could bebserved. People do not seem willing to live near any type ofipeline, although they prefer a CO2 pipeline to a gas pipeline.

The type of plant also plays a role in respondents’ acceptanceudgments. Biogas-fueled power plants are preferred over gas-red ones. This result is in line with earlier studies that showedhat laypeople’s socioeconomic concerns (e.g., about the sustain-bility of energy economies) related to the burning of fossil fuelsay influence their attitudes towards CCS negatively, and more

o than technical concerns about storage safety (Wallquist et al.,009, 2010). In the present study, the power plant type was also

ore important than the place of storage for the acceptance of the

iving situation scenario. Of the three examined elements, storageocation had the smallest influence on acceptance. This is rather sur-rising, given that much of public acceptance research has focused

reenhouse Gas Control 6 (2012) 77–83 81

on the storage part of CCS. Because storage is the most novel ele-ment of CCS, it seems plausible that it would cause the greatestconcern for people. Other research that looked at CO2 storage inisolation found that the public is generally concerned about CCStechnology if a CO2 storage site was to be sited within 5 km of theirhome (European Commission, 2011). In our study, the introduc-tory text put most emphasis on CO2 storage. However, our resultsindicate that the storage part of CCS is less crucial for public per-ception than other parts of the CCS chain. We did indeed find theNIMBY effect for storage location, although less pronounced thanfor pipeline: a storage location in a neighboring canton was pre-ferred over a storage location in one’s own town.

The results of the cluster analysis corroborate the above find-ings and help to paint a more nuanced picture. Three groups ofpeople can be distinguished, which each focus on one elementwhen evaluating CCS scenarios. While “no pipeline” is always thepreferred scenario, there are considerable differences between dif-ferent groups of people. This indicates that the results from theANOVA are not merely an artifact due to the fact that “no pipeline”is such a strong label for the factor level. While cluster 1 showsthe expected strong preference for “no pipeline” scenarios, thereis also a large group of people (cluster 3) that is largely indifferentwhen it comes to the factor pipeline, and their utility scores do notdeviate much from zero. The CCS element that matters most to thatgroup is storage location. Unsurprisingly, they prefer storage of CO2in a neighboring canton over storage in their own town. Interest-ingly, this preference is reversed for cluster 2, which exhibits a smallpreference for storage in their own municipality. This is, however,coupled with a strong preference for CO2 from biogas-fired powerplants as opposed to conventional gas power plants. It is likely thatthe origin of the CO2 is crucial for the acceptance of the storagelocation, together with the absence of pipelines.

This interpretation is supported by the analysis of the interac-tion terms of the ANOVA. In the case where respondents are notaffected by a pipeline, there is a notable NIMBY effect for storagelocation, whereas in situations where they are affected by pipelines,the storage location of the CO2 does not seem to matter much. Thismight seem as if the inclusion of pipelines could be an easy solutionto reduce the NIMBY effect for storage. As the interaction diagram(Fig. 2) shows, however, acceptance levels are a lot lower for sce-narios where there are pipelines present, compared to scenarioswithout pipelines. Field trials for geological storage in densely pop-ulated areas may therefore consider avoiding pipeline transport inorder to increase the likelihood of public acceptance.

The influence of the type of plant producing the CO2 is a differ-ent story. If the CCS system includes a conventional gas-fired plant,respondents preferred storage in a neighboring canton over storagein their own municipality, whereas if the CO2 stems from a biogas-fired power plant, this NIMBY effect seemed to disappear (althoughthe effect size of this interaction is very small). Thus, some peoplemight even be proud of having a BECCS project in their municipality.A reason for this might be the so-called halo effect that describes thefact that people may rate products with one beneficial characteris-tic more favorably on other characteristics, although these are notreally related (Thorndike, 1920). In the case of the type of plant, theterm “bio” may have a positive connotation in terms of naturalnessand healthiness (“bio” being the German translation equivalent of“organic”). Because acceptance of situations that include biogas-fired plants is generally higher than the ones including gas-firedplants, it may, from a perspective of public acceptance, make senseto design field trials for geological storage of CO2 with a supplyof biogenic CO2 instead of CO2 from fossil resources. Field trials

are important, not only because researchers can gain new insightsabout CO2 storage and the properties of a specific storage location,but also because the public can learn about and become more famil-iar with the technology (Singleton et al., 2009). Familiarity is an

8 l of G

ipiltsswaC

ilrniTk

ertcp(pr

aofaowamdcsd

fpwibeCoeta

yesthtatpgebb

Palmgren, C.R., Morgan, G.M., Bruine de Bruin, W., Keith, D.W., 2004. Initial publicperceptions of deep geological and oceanic disposal of carbon dioxide. Environ.

2 L. Wallquist et al. / International Journa

mportant factor in risk perception, as shown by the psychometricaradigm (Fischhoff et al., 1978; Singleton et al., 2009). If familiar-

ty is increased, risk perception shifts with it, generally resulting inower risk perception. BECCS field trials would, under the assump-ion that the biomass production is sustainable, make sense foreveral reasons. Environmental NGOs are more likely to activelyupport the project. In the case of successful storage of CO2, CCSould receive a more climate-friendly framing because of the neg-

tive CO2 emissions created. Moreover, if problems occurred andO2 leaked out, the project would still be climate-neutral.

We think that the present research can give better insightsnto public preferences than evaluating single elements in iso-ation, because the assessment of complete scenarios is moreealistic. Without a given reference, evaluating risks is difficult foron-experts (Hsee, 1996; Hsee and Leclerc, 1998). Therefore, it is

mportant to provide a context when studying public acceptance.his is of particular importance in fields where the public has littlenowledge and where additional communication is needed.

The living scenarios presented to the participants were, how-ver, rather abstract. We are not sure whether, for example, ourespondents realized how much a CO2 plume could spread fromhe injection location. The high importance of pipelines may indi-ate that some people may perceive these as totally new, becauseipelines may be beyond the respondents’ horizon of experiencei.e. they are not aware of existing pipelines). Future research mayresent more extensive and realistic materials, using, for example,eal case studies with maps and more concrete descriptions.

Another limitation of this study is that we could only look at small number of factors and levels. We can therefore not ruleut that other factors are more influential to acceptance than,or example, the NIMBY effect. Another relevant factor for publiccceptance of CCS may be the type of plant operator (e.g., publicr private). Also, the study does not provide evidence to explainhy certain living situations were preferred over others, because

cceptance was the only dependent measure. A complete experi-ental design with more factors and/or factor levels and additional

ependent variables would have resulted in such a large number ofombinations that participants would not have been able to makeense of them anymore and probably would have resulted in manyropouts.

The levels chosen in the present study are particularly validor countries without coal-fired power plants. Because coal-firedower plants are not a realistic option in Switzerland (BFE, 2007),e did not include this type of plant in our scenarios. If it had been

ncluded, the importance of the type of plant would likely have beenigger, because people associate coal-fired power plants with morenvironmental externalities than gas-fired power plants (Europeanommission, 2007). This example shows that the importance scoresf the three factors are determined by the choice of the factor lev-ls. Therefore, the presented importance scores in this study needo be seen in light of the underlying factor levels when conclusionsre drawn.

To conclude, in our study, we used a conjoint analysis and anal-sis of variance to examine relative preferences for CCS systemlements. Our findings indicate that public acceptance of one CCSystem element may depend on the characteristic of another andhat reference information on the whole CCS system, therefore,elps people to evaluate geological storage of CO2. We showedhat it might be possible to avoid the NIMBY effect for field tri-ls of CO2 storage by applying BECCS. We are aware of the manyechnical and economic constraints project developers face whenlanning a field trial for CO2 storage. Nevertheless, our results sug-est that it can be worthwhile for them to examine if technical,conomic, and regulatory frameworks allow public preferences to

e taken into consideration when designing field trials and projectseyond.

reenhouse Gas Control 6 (2012) 77–83

Acknowledgements

The authors would like to thank Swisselectric Research andthe initiative “Carbon Management in Swiss Power Generation”(CARMA), which is funded by CCEM and CCES of the ETH Domain,for supporting this work.

References

Alriksson, S., Öberg, T., 2008. Conjoint analysis. Environ. Sci. Poll. Res. 15, 119-119.Ashworth, P., Boughen, N., Mayhew, M., Millar, F., 2010. From research to action:

now we have to move on CCS communication. Int. J. Greenhouse Gas Control 4,426–433.

Backhaus, K., Erichson, B., Plinke, W., Rolf, W., 2006. Multivariate Analysemethoden:Eine anwendungsorientierte Einführung, 11th ed. Springer, Berlin.

BFE, 2007. Die Energieperspektiven 2035 – Band 1, Synthese. Swiss Federal Officefor Energy (BFE).

Bundesrat, 2011. Bundesrat beschliesst im Rahmen der neuen Energiestrategieschrittweisen Ausstieg aus der Kernenergie (in German) [Federal Councildecides under the new energy strategy, a stepwise phasing out of nuclearenergy].

Carver, R.P., 1990. Reading Rate a Review of Research and Theory. Academic Press,San Diego.

Curry, T., Reiner, D.M., Ansolabehere, S., Herzog, H.J., 2007. A survey of publicattitudes towards climate change and climate change mitigation technolo-gies in the United States: analyses of 2006 results, MIT LFEE 2007-01 WP.http://sequestration.mit.edu/pdf/LFEE 2007 01 WP.pdf.

de Best-Waldhober, M., Daamen, D., Faaij, A., 2009. Informed and uninformed publicopinions on CO2 capture and storage technologies in the Netherlands. Int. J.Greenhouse Gas Control 3, 322–332.

Desbarats, J., Upham, P., Riesch, H., Reiner, D., Brunsting, S., de Best-Waldhober,M., Duetschke, E., Oltra, C., Sala, R., McLachlan, C., 2010. Review of the PublicParticipation Practices for CCS and Non-CCS Projects in Europe. Institute forEuropean Environmental Policy, London.

Devine-Wright, P., 2005. Beyond NIMBYism: towards an integrated frame-work for understanding public perceptions of wind energy. Wind Energy 8,125–139.

Dohle, S., Keller, C., Siegrist, M., 2010. Conjoint measurement of base station sitingpreferences. Hum. Ecol. Risk Assess. 16, 825–836.

Dooley, J.J., Dahowski, R.T., Davidson, C.L., 2009. Comparing existing pipeline net-works with the potential scale of future U.S. CO2 pipeline networks. EnergyProcedia 1, 1595–1602.

Dütschke, E., 2011. What drives local public acceptance – comparing two cases fromGermany. Energy Procedia 4, 6234–6240.

European Commission, 2007. Energy technologies: Knowledge, perception, mea-sures. Special Eurobarometer 262, Wave 65.3 – TNS Opinion & Social.

European Commission, 2011. Public awareness and acceptance of CO2 capture andstorage. Special Eurobarometer 364, Wave 75.1 – TNS Opinion & Social.

Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S., Combs, B., 1978. How safe is safeenough? A psychometric study of attitudes towards technological risks andbenefits. Policy Sci. 9, 127–152.

Fleishman, L.A., De Bruin, W.B., Morgan, M.G., 2010. Informed public preferences forelectricity portfolios with CCS and other low-carbon technologies. Risk Anal. 30,1399–1410.

Frewer, L.J., Howard, C., Hedderley, D., Shepherd, R., 1997. Consumer attitudestowards different food-processing technologies used in cheese production – theinfluence of consumer benefit. Food Qual. Prefer. 8, 271–280.

Green, P.E., Srinivasan, V., 1990. Conjoint-analysis in marketing – new developmentswith implications for research and practice. J. Market. 54, 3–19.

Hair, J.F., 1995. Multivariate Data Analysis with Readings, 4th ed. Prentice-Hall Inter-national, London.

Hsee, C.K., 1996. The evaluability hypothesis: an explanation for preference rever-sals between joint and separate evaluations of alternatives. Behav. Hum. Decis.Process. 67, 247–257.

Hsee, C.K., Leclerc, F., 1998. Will products look more attractive when presentedseparately or together? J. Consum. Res. 25, 175–186.

Itaoka, K., Okuda, Y., Saito, A., Akai, M., 2009. Influential information and factorsfor social acceptance of CCS: the 2nd round survey of public opinion in Japan.Energy Procedia 1, 4803–4810.

L’Orange Seigo, S., Wallquist, L., Dohle, S., Siegrist, M., 2011. Communication ofCCS monitoring activities may not have a reassuring effect on the public. IntJ Greenhouse Gas Control 5, 1674–1679.

Malone, E.L., Dooley, J.J., Bradbury, J.A., 2010. Moving from misinformation derivedfrom public attitude surveys on carbon dioxide capture and storage towardsrealistic stakeholder involvement. Int. J. Greenhouse Gas Control 4, 419–425.

Miller, E., Summerville, J., Buys, L., Bell, L., 2008. Initial public perceptions of carbongeosequestration: implications for engagement and environmental risk com-munication strategies. Int. J. Global Environ. Issue, 147–164.

Sci. Technol. 38, 6441–6450.Read, P., Lermit, J., 2005. Bio-energy with carbon storage (BECS): a sequential deci-

sion approach to the threat of abrupt climate change. Energy 30, 2654–2671.

l of G

S

S

S

SS

S

T

Wolsink, M., 2000. Wind power and the NIMBY-myth: institutional capacity and the

L. Wallquist et al. / International Journa

hackley, S., McLachlan, C., Gough, C., 2005. The public perception of carbon dioxidecapture and storage in the UK: results from focus groups and a survey. Clim.Policy 4, 377–398.

iegrist, M., Stampfli, N., Kastenholz, H., 2009. Acceptance of nanotechnology foods:a conjoint study examining consumers’ willingness to buy. Br. Food J. 111,660–668.

ingleton, G., Herzog, H., Ansolabehere, S., 2009. Public risk perspectives on thegeologic storage of carbon dioxide. Int. J. Greenhouse Gas Control 3, 100–107.

PSS, 2007. SPSS Conjoint 16.0. SPSS Inc., Chicago.teubing, B., Zah, R., Waeger, P., Ludwig, C., 2010. Bioenergy in Switzerland: assessing

the domestic sustainable biomass potential. Renew. Sustain. Energy Rev. 14,2256–2265.

utter, D., Werner, M., Casas, N., Marx, D., Schell, J., Joss, L., Zappone, A., Mazzotti,

M., 2011. Roadmap for testing feasibility of CCS in Switzerland. In: Smart EnergyStrategies Conference, September 21–23, 2011, Zurich, Switzerland.

erwel, B.W., Harinck, F., Ellemers, N., Daamen, D.D.L., 2011. Going beyond the prop-erties of CO2 capture and storage (CCS) technology: how trust in stakeholdersaffects public acceptance of CCS. Int. J. Greenhouse Gas Control 5, 181–188.

reenhouse Gas Control 6 (2012) 77–83 83

Thorndike, E.L., 1920. A constant error in psychological ratings. J. Appl. Psychol. 4,25–29.

Verhagen, M.J.M., 2010. Uitwerking van de afspraken voor de individuele CO2-opslagprojecten die momenteel in voorbereiding zijn (in Dutch) [Elaboration ofthe agreements regarding the individual CO2 storage projects that are currentlyin preparation].

Wallquist, L., Visschers, V.H.M., Siegrist, M., 2009. Lay concepts on CCS deploymentin Switzerland based on qualitative interviews. Int. J. Greenhouse Gas Control 3,652–657.

Wallquist, L., Visschers, V.H.M., Siegrist, M., 2010. Impact of knowledge and mis-conceptions on benefit and risk perception of CCS. Environ. Sci. Technol. 44,6557–6562.

limited significance of public support. Renew. Energy 21, 49–64.Wolsink, M., 2006. Invalid theory impedes our understanding: a critique on

the persistence of the language of NIMBY. Trans. Inst. Br. Geogr. 31,85–91.