- Email: [email protected]
New environmental technology uptake and bias toward the status quo: The case of phytoremediation
Accepted Manuscript New environmental technology uptake and bias toward the status quo: The case of phytoremediation ´ Montpetit, Erick Lachapelle Eric PII: DOI: Reference:
S2352-1864(16)30191-2 http://dx.doi.org/10.1016/j.eti.2016.12.008 ETI 107
To appear in:
Environmental Technology & Innovation
Received date: 11 August 2015 Revised date: 18 October 2016 Accepted date: 27 December 2016 Please cite this article as: Montpetit, ., Lachapelle, E., New environmental technology uptake and bias toward the status quo: The case of phytoremediation. Environmental Technology & Innovation (2016), http://dx.doi.org/10.1016/j.eti.2016.12.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
*Highlights (for review)
Highlights 1. We find a substantial increase in the number of scientific articles documenting the conditions under which phytoremediation provides an effective method to decontaminate soil. 2. We show that practitioners exhibit a bias toward the status quo, even when alternative methods are more suitable. 3. This status quo bias persists even when practitioners are exposed to evidence demonstrating the effectiveness of phytoremediation for particular sites.
New Environmental Technology Uptake and Bias toward the Status Quo: The Case of Phytoremediation
Éric Montpetit (corresponding author) Université de Montréal Department of Political Science [email protected] and
Erick Lachapelle Université de Montréal Department of Political Science [email protected] Université de Montréal Département de science politique CP 6128 succ. Centre-ville Montréal QC Canada H3C 3J7
Paper re-submitted to Environmental Technology and Innovation, March 2016
*Revised Manuscript with No Changes Marked
New Environmental Technology Uptake and Bias toward the Status Quo: The Case of Phytoremediation Abstract A surprisingly large number of sites around the world are left abandoned despite the presence of contaminants posing clear risks for the environment and human health. For many land owners of these sites, the cost of current decontamination technologies is prohibitive. Over the last two decades, discoveries in microbiology and plant science have regularly found that, under certain conditions, phytoremediation is a cost-effective tool for remediating sites with organic contaminants and trace elements. However, awareness and use of this technology by practitioners lags significantly behind that of conventional technology, reflecting a status quo bias and preference for conventional excavation. Drawing on data from an original survey of soil decontamination practitioners, this study sheds new light on why, despite its promise, phytoremediation remains under-used. This research highlights the challenge of transferring scientific knowledge from the laboratory to practitioners working to mitigate serious environmental problems. Keywords Phytoremediation; soil decontamination; status quo bias; learning; technology uptake. 1. Introduction Many scientists work toward developing solutions to the world’s pressing environmental problems in the hope that their discoveries will prompt appropriate responses. The use of new environmental solutions by practitioners, however, does not always follow a straightforward logic. Indeed, technological transfer from developers to end-users is influenced by a variety of factors that determine the successful uptake of new technologies.1 Beyond relative costs, these include institutional factors, like increasing returns, organizational demands for certainty and protection against liability, as well as cognitive biases, like familiarity and habit, that can “lock-in” technologies even when more efficient ones are available.2 This bias toward the status quo presents an important challenge for the integration of new scientific findings and technologies into the day-to-day behavior of practitioners on the ground. Looking at the field of soil contamination, we show that decontamination professionals demonstrate a bias toward using conventional methods. We also show that exposure to clear cues regarding the relative performance of alternative technologies can alter this bias, if only slightly. While we acknowledge that there are many reasons for this status quo bias, we focus here on the influence of experience and familiarity with conventional technologies
1
as a factor leading professionals to avoid alternative technologies, even when faced with clear evidence that a new technology offers a more efficient option in specified conditions. These conclusions are supported by a survey of 113 decontamination experts who display a marked sense of apathy toward the use of plants and other organisms to remediate contaminated sites, despite substantial scientific evidence published in the last 20 years demonstrating the effectiveness of these phytoremediation methods in a range of conditions. In fact, the scientific literature presents a large number of experiments in which trees, plant species and related microorganisms were shown to effectively degrade several organic contaminants, stabilize toxic metals in insoluble form and even phytoextract other trace elements from contaminated soils. In an experiment embedded in a practitioner survey, we exposed remediation experts to clear and credible evidence demonstrating the effectiveness of willows at reducing heavy metals in soil. Despite exposure to this scientific research, we found only a moderate change in the experts’ attitudes toward phytoremediation. Our results indicate that preferences for familiar approaches among front line professionals are difficult to change. Scientific work certainly can change attitudes, but only gradually and over the long term. 2. Context and Theory Soil contamination is a serious global environmental problem: in Europe alone, there are an estimated 160,000 known sites where pollutants present clear risks for humans and ecosystems.3 Government estimates suggest the problem is no less serious in the United States, which currently has 1,438 abandoned sites on its National Priority List, comprised of only the “worst hazardous waste sites.”4 To make matters worse, the exact number of contaminated sites is unknown. Panagos et al.3 suggest there are over a million contaminated sites in Europe while McIntyre5 estimates the figure at 500,000 for the United States. The toxicity of the substances found on these sites varies, but several of them are known human carcinogens, including trichloroethylene, tetrachloreothylene, arsenic, benzene, cadmium, chromium, and polychlorinated biphenyl (PCB).6 The populations directly exposed to contaminated sites are thus at risk of developing serious health problems. Moreover, it has been shown that populations residing near contaminated sites are at statistically higher risk of dying from some cancers, notably leukemia, urinary bladder and gastrointestinal cancers, than the rest of the population.6 Contaminants also find their way into the food chain through soil leaching, wind erosion, plant uptake and other natural processes, thus exposing relatively remote populations to health risks.7,8 The risks associated with site contamination have been deemed serious enough to spur the American Environmental Protection Agency to develop a program in 1980, known as the Superfund, to remediate the most dangerous sites.9 Nevertheless, most contaminated sites today are still left abandoned, owing to the at times prohibitive cost of conventional 2
excavation methods for removing pollutants from contaminated areas and transporting them to landfills or treatment facilities off-site.10 In this context, many scientists have been working to develop cost-effective methods that can significantly reduce risks, and in some cases entirely remediate contaminated sites. Over the last 20 years, phytoremediation has emerged as a promising method. In fact, some scientists have identified several new tree and plant cultivars capable of mitigating soil and water pollution. Knowledge on the effect of these cultivars is sufficiently advanced to apply phytoremediation with predictable success in a variety of soil conditions.11 Moreover, some economists have shown not only that trees and plants offer a remediation approach that is less expensive than engineered cleanup technologies to meet remedial goals, but have also shown that planting and maintenance costs of these trees and plants can be recovered by revenues generated from their harvested biomass.12,13 Phytoremediation thus has a low environmental impact, requiring a minimum of equipment while avoiding the transportation of large amounts of soil, typical of engineered cleanup technologies. Although research into the remediation potential of plants can be traced back to the 1970s, the earliest article containing the word phytoremediation in its title and found in Thomson Reuters’ Web of Science database was published in 1994. Subsequently, publication of phytoremediation research increased substantially, beginning with a review by a team of researchers led by Salt,14 which demonstrated the ability of plants to extract and stabilize toxic heavy metals. The review generated significant interest among scientists, and is cited 972 times by other articles found in Thomson’ Reuters’ Web of Science as of February 2016. The most cited publication on the topic, authored by another of Salt’s teams, received 1229 citations as of the same date.10 The Web of Science “Science Citation Index” includes 1795 research articles and reviews published between 1994 and 2015 with the word phytoremediation in the title, not counting the numerous books, book chapters, editorial material and conference proceedings, nor the articles using related words such as phytostabilisation or phytoextraction in their title. As shown in Figure 1, the publication of research on phytoremediation has gained significant popularity over the past 20 years. We interpret this as a substantial increase in scientific interest in the area of phytoremediation. Moreover, our review of this literature suggests there is substantial evidence the technology works under specified conditions.
3
Figure 1: Evolution in the number of articles and reviews with the word phytoremediation in their title
Data source: Thomson Reuters’ Web of Science. The content of these articles suggests that phytoremediation is effective in an array of conditions. Take the 11 reviews9,11,15–23 with the word phytoremediation in their title that are listed in the Web of Science for 2013. Each one aims at promoting a promising avenue (e.g. a new bacteria,22 a new plant species,19 a transgenic manipulation,17 and even electrokinetics15) to improve the efficacy of phytoremediation in particularly difficult sites. Each review also begins with a presentation of pot and field experiments showing the effectiveness of current phytoremediation practices in suitable conditions. One review11 is somewhat critical of the procedure, insisting that between 30% and 40% of the contaminated sites in the United States are characterized by the presence of both organic and inorganic contaminants, and stating that such a mix of pollutants can be difficult to decontaminate with phytoremediation, owing to heavy metals’ negative interaction with the microorganisms known to degrade organic contaminants. The authors nonetheless cite reports of successful phytoremediation on sites with both organic and inorganic contaminants, indicating that phytoremediation can be useful on some sites with mixed pollutants. They further argue that determining the extent of this usefulness requires better knowledge of microbial activities in plant rhizospheres. More importantly, they conclude that “It is evident that phytoremediation provides a realistic and sustainable option for removing organic pollutants from contaminated sites”.11 While more prudent with inorganics, they provide a list of plant species (notably of Salix, Brassica and Populus), 4
which were shown to accumulate Cu, Zn, Ni, Cd and Pb in their roots and/or shoots and to stabilize As, Co, Cu, Pb and Zn in soil. This growing body of research has thus raised the scientific profile of phytoremediation while highlighting the conditions under which it is most effectively applied. Despite the increase in scientific interest for these methods, practitioners working in the area of soil decontamination have a relatively low level of knowledge about phytoremediation in general.24 This low level of knowledge among soil decontamination practitioners is related to a preference in favor of conventional methods, like excavation, for the remediation of soil.25 In this article, we explore whether this preference for conventional methods is suggestive of a status quo bias,26 that is, a preference to stick with a previous technological decision even when a superior option is presented. We also explore the strength of this bias by examining the extent to which the preference for conventional methods is maintained even when decision-makers are presented with new information nudging them toward an alternative, more efficient option. To be sure, the decision on which environmental technology to select for a given remediation problem is based on numerous factors, including cost (i.e. whether cheaper alternatives are available), time sensitivity (i.e. how quickly the remediation occurs relative to project imperatives), certainty and liability (i.e. the efficacy of remediation relative to specific targets and objectives) and the broader risk evaluation framework employed (i.e. whether the objective is to manage or eliminate risks from contamination altogether). It is also true that the efficacy of phytoremediation is limited to particular contexts, and that the performance of this method is compromised by such factors as level and depth of contamination, the type of contaminants present, as well as the length of time it takes to decontaminate affected areas so that they reach specified acceptable levels.11 However, a growing body of research has found that phytoremediation works effectively with organic contaminants at relatively low concentrations, and that performance is also effective, – if slower and at times helped by other technologies – for decontaminating trace elements.15,16,18 In cases where phytoremediation offers several advantages over conventional excavation, do practitioners nevertheless show a preference for conventional methods? If so, how deep is this bias for the status quo? 3. Methods and Results To answer these questions, we administered an online survey to the 193 decontamination experts in Quebec, Canada, between November 18 and December 12, 2013. Like other soil decontamination practitioners around the world, these experts possess relevant qualifications in the field of soil remediation, and are accredited with the professional credentials to oversee remediation projects. As a result, these experts also have substantial 5
influence over which remediation technologies are used. Of the 193 experts who received an invitation to complete the survey, 113 (or 59%) filled out the questionnaire partially, while 94 (or 49%) completed it entirely. The survey comprised a short knowledge test on phytoremediation, based on four simple true or false questions: phytoremediation involves plants (true)? Phytoremediation always requires harvesting (false)? Phytoextraction removes organic contaminants (false)? Some plants can extract inorganic pollutants while simultaneously degrading organic contaminants (true)? Any single review published on phytoremediation in the last twenty years contains sufficient information to obtain a perfect score on the test. Yet the mean test score obtained by our experts was 2.18 out of 4 (54%), with a standard deviation of 1.09. This relatively poor score is consistent with the respondents’ self-assessed level of phytoremediation knowledge, with 73% of the sample admitting to having below-average knowledge on this technology. In other words, the survey provides evidence that the large quantity of scientific knowledge produced on phytoremediation since 1995 has failed to reach experts on the ground. Moreover, we find that decontamination professionals working in Quebec show little inclination to recommend phytoremediation, even on parcels of land where scientific research has found this method to be appropriate. For instance, we provided respondents with the description of a site with light and shallow mixed contamination owned by a municipality unwilling to spend much on remediation due to a lack of immediate plans for the area. More importantly, the characterization of the soil provided in the description matched that of soil that has been successfully decontaminated with phytoremediation. 27 We thus loosely based our hypothetical scenario on site characteristics and decontamination profile that made phytoremediation an ideal solution for this particular site. We then asked experts to rate on a 0-6 Likert scale the acceptability of a conventional remediation plan, that is, a plan involving excavation and off-site treatment of the soil. On the same survey vignette, we also asked experts to rate a plan to decontaminate the site with willows and poplars, two of the best-known accumulators. Comparing levels of expert acceptability between the conventional and phytoremediation options, we find that experts are significantly more likely to endorse the former, despite the scientifically documented advantages of phytoremediation in this particular scenario. An un-paired t-test was conducted, revealing a significant difference in the scores for excavation and treatment offsite (M=4.43, SD=1.28) and phytoremediation (M=2.89, SD=1.53); t(220), p=0.00001. Overall, we find that 31% of respondents indicated phytoremediation to be acceptable (i.e. assigned a score of over 3), while only 19% preferred it to costly excavation and off-site treatment (i.e. assigned a higher score to phytoremediation than to excavation). This is true despite the fact that, in this particular instance, phytoremediation presents an easily implementable, low-cost and effective solution for this particular site.
6
To test whether this low acceptability of phytoremediation could be explained by the nature of the site’s contaminants, we presented respondents with a second scenario. In this case, half of the respondents were presented with a site characterized by the light presence of organic substances only, while the other half was assigned a site lightly polluted with trace elements. As in the first scenario, these cases were loosely based on existing research characterizing instances of successful remediation using plants at a given level of contamination. However, relative to trace elements, remediation of organic contaminants with phytoremediation has produced stronger results.11 Respondents exposed to the scenario with organic contaminants might thus be expected to show higher levels of acceptability. Basing our site characterization on this, we found the mean level of acceptance rose to above 3 (on the same 6 point Likert scale) in each case, indicating a higher level of acceptability and perhaps suggesting some acquiescence bias as respondents learned that the survey was on phytoremediation. In any event, we note that this level of acceptability was still far below that observed for excavation and treatment off site (M=4.43, SD=1.28). Moreover, we failed to find a statistically significant difference between mean levels of acceptance across the organic contaminants (M=3.80, SD=1.27) and trace elements (M=3.61, SD 1.50) conditions. We thus find evidence of a status quo bias that appears to systematically work against the uptake of phytoremediation, regardless of the type of contaminants found on site. To make sense of the clear preference for excavation expressed by respondents, we ran a logistic regression and found, after controlling for knowledge and other confounding factors, that the frequency with which experts have recommended excavation in their recent practice decreases the likelihood of preferring phytoremediation. This is exactly what the status quo bias would predict—those having more experience with conventional methods should be more likely to maintain a preference for this method, even when presented with a relatively attractive alternative. As shown in Figure 2, the probability of preferring phytoremediation drops from 0.5 to 0.2 for experts who have recommended excavation more than ten times in recent years. This difference is statistically significant. In other words, accustomed to excavation, experts display a status quo bias, even when controlling for self-assessed knowledge of phytoremediation. Exactly how far does this bias go? Is it resistant to new information regarding the suitability of phytormediation under specific conditions?
7
Figure 2: the effect of recommending excavation more or less frequently on the probability of preferring phytoremediation
Difference significant at p=0.02. To understand the extent to which new scientific research can curb this status quo bias, we embedded an experiment within our survey. We exposed a random subsample of half of the respondents to the abstract of a fictional scientific article presenting unambiguous evidence that phytoremediation using two species of willows will effectively remove Zn, Cu and Pb from polluted soil. To maximize the effect of the treatment, we indicated that the article was from a 2010 issue of Nature and authored by MIT professors. The respondents forming the control group were not exposed to this abstract but, like the members of the experimental group, were asked to re-read the site description presented at the very beginning of the survey (and mentioned above, as the first scenario). Respondents were thus asked to provide a second rating of the phytoremediation plan originally presented in the first scenario they were presented. Crucially, this scenario was characterized by the presence of Zn, Cu and Pb, as well as a remediation plan using willows—just as that featured in the abstract of the fictitious MIT study. Details of this experiment can be found in reference 28. As shown in Figure 3, the subjects who read the article’s abstract increased their acceptability of phytoremediation. The mean acceptability score in the control group is 3.26, compared with 3.81 in the experimental group. This difference proved significant at p=0.05 in a statistical t test. Just to put things into perspective, however, the mean 8
acceptability rate of the conventional plan was 4.4. In other words, exposing on-the-ground experts to scientific knowledge makes a difference, albeit a relatively small one. This reluctance to integrate new information into the assessment of decontamination plans is indicative of a status quo bias that works against the uptake of phytoremediation by decontamination professionals in Quebec. Figure 3: Difference in the acceptability of phytoremediation between control and experimental groups
Difference significant at p=0.05. A careful examination of the reasons given by survey respondents to justify their evaluation of the phytoremediation plan, in conjunction with the justification given for their evaluation of the conventional remediation plan, further supports the status quo bias thesis. In fact, after respondents evaluated each, they were invited to justify their evaluation of each remediation plan. Those who approved phytoremediation wrote comments similar to the following: “Given that time isn’t an issue, this is the most environmentally friendly and least costly of the methods”. This respondent also indicated that excavation and off-site treatment were “too expensive and unnecessary.” Conversely, several respondents who most forcefully rejected phytoremediation justified their evaluation with statements that reveal unreasoned principled objection, an attitude to expect from individuals who are unwilling to consider alternatives to practices to which they are accustomed. One wrote: “this is utopia.” Another simply stated: “this thing doesn’t work.” Meanwhile these same
9
respondents simultaneously affirmed, “excavating the soil is the proper way to go.” Some less vehemently opposed respondents, who nonetheless opposed phytoremediation invoked reasons, most notably time, which made little sense in the context of the specific scenario with which they were presented. Perhaps these respondents had not read the site description carefully enough, but they are perhaps also opposed in principle and simply gave a common generic reason to justify their overly critical evaluation of phytoremediation. Supporting this latter explanation is the fact that several of these respondents generally failed to provide specific reasons to support excavation. Other justifications provide even more straightforward evidence of a status quo bias. One respondent said the method is “not popular enough among fellow experts” to be approved by him. Another one feared that “bankers and public authorities are not familiar enough with the method” to justify advising a client to use phytoremediation. These two respondents also wrote that excavation was acceptable because it is the “classic” way to rehabilitate sites. All of this is consistent with the status quo bias, which is one important barrier to the uptake of new technology in the field of soil remediation. 4. Discussion What do these findings say about the uptake of new environmental technologies in the area of soil decontamination? We find considerable evidence of a status quo bias, working in favor of conventional methods, like excavation and treatment off site, even in cases where phytoremediation presents clear advantages. In fact, we show that respondents most opposed to phytoremediation are those who have recommended conventional off-site treatment methods most frequently, even when controlling for knowledge of phytoremediation. When examined in conjunction with the responses obtained from the open-ended text boxes, our findings suggest that a status quo bias characterizes soil decontamination professionals, tempering the adoption of phytoremediation on sites for which the technology is well suited. While a preference for conventional methods is not evidence in and of itself that such a bias exists, the expression of principled objections— held despite the suitability of phytoremediation to some of the scenarios we presented to practitioners—indicates an evident bias for the status quo. In addition, we find that this status quo bias is relatively “sticky,” and exposure to new scientific knowledge is unlikely to break this decision-making inertia in the uptake of remediation technology. To be sure, our experimental manipulation involved exposing practitioners to the abstract of a single scientific publication. Though we might expect the effect of such limited exposure to be modest, the increase in support for that we do observe does little to close the gap in preferences between phytoremediation and conventional methods. The fact that new information does not increase acceptability to a similar level
10
suggests practitioners are reluctant to update their prior judgments and instead fall back on methods with which they are most familiar. Finally, there are a number of reasons why practitioners might prefer conventional methods to phytoremediation. Beyond a cognitive bias, practitioners might require certainty of outcomes—for a variety of reasons, including liability, for instance—that tilt the calculus in favor of well-known, effective strategies, like excavation. The institutional setting in which practitioners are embedded—with insurers, banks, and even government policy—might also work to make conventional excavation the preferred approach. It is possible that professionals have vested interests in particular technologies, either because excavation services are provided by their firms, or because of institutionalized working relationships that have developed in this sector over the years. While we do not deny that such factors play a role, they are beyond the scope of this particular study. To a certain extent, we also control for some of these factors, by comparing across scenarios that are more and less conducive to phytoremediation. The fact that we find little difference across scenarios, regardless of context, suggests there is an entrenched preference for conventional methods, which we argue is consistent with a status quo bias in the selection of appropriate remediation technology in the Canadian province of Quebec.
5. Conclusion Individuals’ daily behavior is comprised of decisions based on simple cognitive rules and conventions, whether these individuals are experts, politicians or ordinary citizens. 29 Indeed, it would be unreasonable to expect individuals to undertake a careful examination of scientific knowledge every time they had to make a decision, even when they are qualified to read the scientific literature. Individuals are economical and prudent in making decisions. As a consequence, they prefer options closest to their experience over options that require extensive research.30 Such a state of affairs induces a bias that favors the status quo. This cognitive bias encourages and reinforces the type of “lock-in” described by scholars of technology adoption, and which we have observed here in the field of soil decontamination.2 Despite this bias for the status quo in the uptake of new environmental technology, we have shown that—in the case of phytoremediation—scientific knowledge can begin to displace conventions and thereby influence decisions, if only gradually. Scientific knowledge surely does not reach the ground in a straightforward manner (that is, when it does at all), but rather begins to be incorporated into practice through small incremental changes in attitudes. The experimental results discussed here can be viewed as a small but important step in this process, which can improve acceptability at the margin. Overall, they suggest 11
that scientific knowledge can make a difference, and we might expect that the accumulation of scientific research might slowly shift the views of decontamination experts and quite possibly those working in other environmental problem areas as well. Over time, gradual acceptance of new technologies and practices—like phytoremediation—are likely to be influenced by the accumulation of scientific evidence and its communication to practitioners at a level of detail that clearly indicates the benefits, limits and general advantages of the new method relative to those conventionally used. In the case of phytoremediation, this may result in full or partial remediation of large parcels of polluted land currently left abandoned. As the case of phytoremediation suggests, status quo biases stemming from cognitive limitations are likely to continue to pose a barrier to the uptake of new technologies. Awareness of such biases is therefore an important first step in mitigating their impacts. Among practitioners, this means being aware of one’s conventions and being open to learning about alternative methods with some promise. Among scientists, this may imply increasing the effort to disseminate detailed findings in a variety of publication venues. Indeed, our research suggests publication in scientific journals may not be the best way to reach practitioners working on the ground. Success in decontaminating a site with phytoremediation depends on specialized knowledge about the conditions under which the technology functions best. These include the nature of the contaminants, levels of contamination as well as contamination depth. Detailing the strengths and limits of new techniques in a variety of settings is essential when communicating to practitioners who have a variety of criteria in mind when selecting among alternative technologies. Overall, we find that knowledge is key to the successful implementation of a phytoremediation plan on a contaminated site, even among practitioners who are already convinced by the technology’s merits. Reaching professionals on the ground is therefore doubly important: while it encourages acceptance, it also contributes to implementation success.
Acknowledgement The authors acknowledge the financial support of Genome Quebec, Genome Canada and the Social Sciences and Humanities Research Council of Canada. They also thank the members of Genorem for their contribution to this research, as well as Monika Smaz and Irena Nedeva who provided excellent research assistance.
References 1.
Roupas, P. Human and Organisational Factors Affecting Technology Uptake by 12
Industry. Innov. Manag. Policy Pract. 10, 4–28 (2008). 2. Arthur, W. B. Positive Feedbacks in the Economv. Sci. Am. 262, 92–99 (1990). 3. Panagos, P., Van Liedekerke, M., Yigini, Y. & Montanarella, L. Contaminated Sites in Europe: Review of the Current Situation Based on Data Collected through a European Network. J. Environ. Public Health 2013, 1–11 (2013). 4. Search for Superfund Sites Where You Live | Superfund | US EPA. Available at: http://www.epa.gov/superfund/search-superfund-sites-where-you-live. (Accessed: 22nd February 2016) 5. McIntyre, T. in Phytoremediation (ed. Tsao, D.) 78, 97–123 (Springer Berlin Heidelberg, 2003). 6. Johnson, B. L. & DeRosa, C. The Toxicologic Hazard of Superfund Hazardous-Waste Sites. Rev. Environ. Health 12, (1997). 7. Chou, C. L., Paon, L. A. & Moffatt, J. D. Cadmium, Copper, Manganese, Silver, and Zinc in Rock Crab ( Cancer irroratus ) from Highly Copper Contaminated Sites in the Inner Bay of Fundy, Atlantic Canada. Bull. Environ. Contam. Toxicol. 68, 885–892 (2002). 8. Barrento, S. et al. Essential Elements and Contaminants in Edible Tissues of European and American Lobsters. Food Chem. 111, 862–867 (2008). 9. Ali, H., Khan, E. & Sajad, M. A. Phytoremediation of Heavy Metals: Concepts and Applications. Chemosphere 91, 869–881 (2013). 10. Salt, D. E., Smith, R. D. & Raskin, I. Phytoremediation. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 643–668 (1998). 11. Batty, L. C. & Dolan, C. The Potential Use of Phytoremediation for Sites With Mixed Organic and Inorganic Contamination. Crit. Rev. Environ. Sci. Technol. 43, 217–259 (2013). 12. Thewys, T., Witters, N., Meers, E. & Vangronsveld, J. Economic Viability of Phytoremediation of a Cadmium Contaminated Agricultural Area Using Energy Maize. Part II: Economics of Anaerobic Digestion of Metal Contaminated Maize in Belgium. Int. J. Phytoremediation 12, 663–679 (2010). 13. Lewandowski, I., Schmidt, U., Londo, M. & Faaij, A. The Economic Value of the Phytoremediation Function – Assessed by the Example of Cadmium Remediation by Willow (Salix ssp). Agric. Syst. 89, 68–89 (2006). 14. Salt, D. E. et al. Phytoremediation: A Novel Strategy for the Removal of Toxic Metals from the Environment Using Plants. Nat. Biotechnol. 13, 468–474 (1995). 15. Cameselle, C., Chirakkara, R. A. & Reddy, K. R. Electrokinetic-Enhanced Phytoremediation of Soils: Status and Opportunities. Chemosphere 93, 626–636 (2013). 16. Leung, H.-M. et al. Interactions Between Arbuscular Mycorrhizae and Plants in Phytoremediation of Metal-Contaminated Soils: A Review. Pedosphere 23, 549–563 (2013). 17. Lee, J. H. An Overview of Phytoremediation As a Potentially Promising Technology for Environmental Pollution Control. Biotechnol. Bioprocess Eng. 18, 431–439 (2013). 18. Gupta, D. K., Huang, H. G. & Corpas, F. J. Lead Tolerance in Plants: Strategies for Phytoremediation. Environ. Sci. Pollut. Res. 20, 2150–2161 (2013). 19. Lokhande, V. H., Gor, B. K., Desai, N. S., Nikam, T. D. & Suprasanna, P. Sesuvium Portulacastrum: a Plant for Drought, Salt Stress, Sand Fixation, Food and Phytoremediation. A Review. Agron. Sustain. Dev. 33, 329–348 (2013). 20. Agostini, E., Talano, M. A., González, P. S., Oller, A. L. W. & Medina, M. I. Application of Hairy Roots for Phytoremediation: What Makes them an Interesting Tool for this Purpose? Appl. Microbiol. Biotechnol. 97, 1017–1030 (2013). 13
21. Kanu, S. A., Okonkwo, J. O. & Dakora, F. D. Aspalathus Linearis (Rooibos tea) as Potential Phytoremediation Agent: a Review on Tolerance Mechanisms for Aluminum Uptake. Environ. Rev. 21, 85–92 (2013). 22. Tak, H. I., Ahmad, F. & Babalola, O. O. in Reviews of Environmental Contamination and Toxicology Volume 223 (ed. Whitacre, D. M.) 223, 33–52 (Springer New York, 2013). 23. Rathod, P. H., Rossiter, D. G., Noomen, M. F. & van der Meer, F. D. Proximal Spectral Sensing to Monitor Phytoremediation of Metal-Contaminated Soils. Int. J. Phytoremediation 15, 405–426 (2013). 24. Lachapelle, E. & Montpetit, E. Putting Phytoremediation into Action. The Scientist 29, 18–19 (2015). 25. Montpetit, É. & Lachapelle, E. Information, Values and Expert Decision-Making: the Case of Soil Decontamination. Policy Sci. (2015). doi:10.1007/s11077-015-9225-x 26. Samuelson, W. & Zeckhauser, R. Status Quo Bias in Decision Making. J. Risk Uncertain. 1, 7–59 (1988). 27. Doyle, C. La Phytoremédiation: une solution à la contamination des sites de traitement du bois? (Université de Sherbrooke, 2008). 28. Montpetit, E. & Lachapelle, E. Can Policy Actors Learn from Academic Scientists? Environ. Polit. 24, 661–680 (2015). 29. Kahneman, D. Maps of Bounded Rationality: Psychology for Behavioral Economics. Am. Econ. Rev. 93, 1449–1475 (2003). 30. Lindblom, C. E. The Science of ‘Muddling Through’. Public Adm. Rev. 19, 79–88 (1959).
14
S2352-1864(16)30191-2 http://dx.doi.org/10.1016/j.eti.2016.12.008 ETI 107
To appear in:
Environmental Technology & Innovation
Received date: 11 August 2015 Revised date: 18 October 2016 Accepted date: 27 December 2016 Please cite this article as: Montpetit, ., Lachapelle, E., New environmental technology uptake and bias toward the status quo: The case of phytoremediation. Environmental Technology & Innovation (2016), http://dx.doi.org/10.1016/j.eti.2016.12.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
*Highlights (for review)
Highlights 1. We find a substantial increase in the number of scientific articles documenting the conditions under which phytoremediation provides an effective method to decontaminate soil. 2. We show that practitioners exhibit a bias toward the status quo, even when alternative methods are more suitable. 3. This status quo bias persists even when practitioners are exposed to evidence demonstrating the effectiveness of phytoremediation for particular sites.
New Environmental Technology Uptake and Bias toward the Status Quo: The Case of Phytoremediation
Éric Montpetit (corresponding author) Université de Montréal Department of Political Science [email protected] and
Erick Lachapelle Université de Montréal Department of Political Science [email protected] Université de Montréal Département de science politique CP 6128 succ. Centre-ville Montréal QC Canada H3C 3J7
Paper re-submitted to Environmental Technology and Innovation, March 2016
*Revised Manuscript with No Changes Marked
New Environmental Technology Uptake and Bias toward the Status Quo: The Case of Phytoremediation Abstract A surprisingly large number of sites around the world are left abandoned despite the presence of contaminants posing clear risks for the environment and human health. For many land owners of these sites, the cost of current decontamination technologies is prohibitive. Over the last two decades, discoveries in microbiology and plant science have regularly found that, under certain conditions, phytoremediation is a cost-effective tool for remediating sites with organic contaminants and trace elements. However, awareness and use of this technology by practitioners lags significantly behind that of conventional technology, reflecting a status quo bias and preference for conventional excavation. Drawing on data from an original survey of soil decontamination practitioners, this study sheds new light on why, despite its promise, phytoremediation remains under-used. This research highlights the challenge of transferring scientific knowledge from the laboratory to practitioners working to mitigate serious environmental problems. Keywords Phytoremediation; soil decontamination; status quo bias; learning; technology uptake. 1. Introduction Many scientists work toward developing solutions to the world’s pressing environmental problems in the hope that their discoveries will prompt appropriate responses. The use of new environmental solutions by practitioners, however, does not always follow a straightforward logic. Indeed, technological transfer from developers to end-users is influenced by a variety of factors that determine the successful uptake of new technologies.1 Beyond relative costs, these include institutional factors, like increasing returns, organizational demands for certainty and protection against liability, as well as cognitive biases, like familiarity and habit, that can “lock-in” technologies even when more efficient ones are available.2 This bias toward the status quo presents an important challenge for the integration of new scientific findings and technologies into the day-to-day behavior of practitioners on the ground. Looking at the field of soil contamination, we show that decontamination professionals demonstrate a bias toward using conventional methods. We also show that exposure to clear cues regarding the relative performance of alternative technologies can alter this bias, if only slightly. While we acknowledge that there are many reasons for this status quo bias, we focus here on the influence of experience and familiarity with conventional technologies
1
as a factor leading professionals to avoid alternative technologies, even when faced with clear evidence that a new technology offers a more efficient option in specified conditions. These conclusions are supported by a survey of 113 decontamination experts who display a marked sense of apathy toward the use of plants and other organisms to remediate contaminated sites, despite substantial scientific evidence published in the last 20 years demonstrating the effectiveness of these phytoremediation methods in a range of conditions. In fact, the scientific literature presents a large number of experiments in which trees, plant species and related microorganisms were shown to effectively degrade several organic contaminants, stabilize toxic metals in insoluble form and even phytoextract other trace elements from contaminated soils. In an experiment embedded in a practitioner survey, we exposed remediation experts to clear and credible evidence demonstrating the effectiveness of willows at reducing heavy metals in soil. Despite exposure to this scientific research, we found only a moderate change in the experts’ attitudes toward phytoremediation. Our results indicate that preferences for familiar approaches among front line professionals are difficult to change. Scientific work certainly can change attitudes, but only gradually and over the long term. 2. Context and Theory Soil contamination is a serious global environmental problem: in Europe alone, there are an estimated 160,000 known sites where pollutants present clear risks for humans and ecosystems.3 Government estimates suggest the problem is no less serious in the United States, which currently has 1,438 abandoned sites on its National Priority List, comprised of only the “worst hazardous waste sites.”4 To make matters worse, the exact number of contaminated sites is unknown. Panagos et al.3 suggest there are over a million contaminated sites in Europe while McIntyre5 estimates the figure at 500,000 for the United States. The toxicity of the substances found on these sites varies, but several of them are known human carcinogens, including trichloroethylene, tetrachloreothylene, arsenic, benzene, cadmium, chromium, and polychlorinated biphenyl (PCB).6 The populations directly exposed to contaminated sites are thus at risk of developing serious health problems. Moreover, it has been shown that populations residing near contaminated sites are at statistically higher risk of dying from some cancers, notably leukemia, urinary bladder and gastrointestinal cancers, than the rest of the population.6 Contaminants also find their way into the food chain through soil leaching, wind erosion, plant uptake and other natural processes, thus exposing relatively remote populations to health risks.7,8 The risks associated with site contamination have been deemed serious enough to spur the American Environmental Protection Agency to develop a program in 1980, known as the Superfund, to remediate the most dangerous sites.9 Nevertheless, most contaminated sites today are still left abandoned, owing to the at times prohibitive cost of conventional 2
excavation methods for removing pollutants from contaminated areas and transporting them to landfills or treatment facilities off-site.10 In this context, many scientists have been working to develop cost-effective methods that can significantly reduce risks, and in some cases entirely remediate contaminated sites. Over the last 20 years, phytoremediation has emerged as a promising method. In fact, some scientists have identified several new tree and plant cultivars capable of mitigating soil and water pollution. Knowledge on the effect of these cultivars is sufficiently advanced to apply phytoremediation with predictable success in a variety of soil conditions.11 Moreover, some economists have shown not only that trees and plants offer a remediation approach that is less expensive than engineered cleanup technologies to meet remedial goals, but have also shown that planting and maintenance costs of these trees and plants can be recovered by revenues generated from their harvested biomass.12,13 Phytoremediation thus has a low environmental impact, requiring a minimum of equipment while avoiding the transportation of large amounts of soil, typical of engineered cleanup technologies. Although research into the remediation potential of plants can be traced back to the 1970s, the earliest article containing the word phytoremediation in its title and found in Thomson Reuters’ Web of Science database was published in 1994. Subsequently, publication of phytoremediation research increased substantially, beginning with a review by a team of researchers led by Salt,14 which demonstrated the ability of plants to extract and stabilize toxic heavy metals. The review generated significant interest among scientists, and is cited 972 times by other articles found in Thomson’ Reuters’ Web of Science as of February 2016. The most cited publication on the topic, authored by another of Salt’s teams, received 1229 citations as of the same date.10 The Web of Science “Science Citation Index” includes 1795 research articles and reviews published between 1994 and 2015 with the word phytoremediation in the title, not counting the numerous books, book chapters, editorial material and conference proceedings, nor the articles using related words such as phytostabilisation or phytoextraction in their title. As shown in Figure 1, the publication of research on phytoremediation has gained significant popularity over the past 20 years. We interpret this as a substantial increase in scientific interest in the area of phytoremediation. Moreover, our review of this literature suggests there is substantial evidence the technology works under specified conditions.
3
Figure 1: Evolution in the number of articles and reviews with the word phytoremediation in their title
Data source: Thomson Reuters’ Web of Science. The content of these articles suggests that phytoremediation is effective in an array of conditions. Take the 11 reviews9,11,15–23 with the word phytoremediation in their title that are listed in the Web of Science for 2013. Each one aims at promoting a promising avenue (e.g. a new bacteria,22 a new plant species,19 a transgenic manipulation,17 and even electrokinetics15) to improve the efficacy of phytoremediation in particularly difficult sites. Each review also begins with a presentation of pot and field experiments showing the effectiveness of current phytoremediation practices in suitable conditions. One review11 is somewhat critical of the procedure, insisting that between 30% and 40% of the contaminated sites in the United States are characterized by the presence of both organic and inorganic contaminants, and stating that such a mix of pollutants can be difficult to decontaminate with phytoremediation, owing to heavy metals’ negative interaction with the microorganisms known to degrade organic contaminants. The authors nonetheless cite reports of successful phytoremediation on sites with both organic and inorganic contaminants, indicating that phytoremediation can be useful on some sites with mixed pollutants. They further argue that determining the extent of this usefulness requires better knowledge of microbial activities in plant rhizospheres. More importantly, they conclude that “It is evident that phytoremediation provides a realistic and sustainable option for removing organic pollutants from contaminated sites”.11 While more prudent with inorganics, they provide a list of plant species (notably of Salix, Brassica and Populus), 4
which were shown to accumulate Cu, Zn, Ni, Cd and Pb in their roots and/or shoots and to stabilize As, Co, Cu, Pb and Zn in soil. This growing body of research has thus raised the scientific profile of phytoremediation while highlighting the conditions under which it is most effectively applied. Despite the increase in scientific interest for these methods, practitioners working in the area of soil decontamination have a relatively low level of knowledge about phytoremediation in general.24 This low level of knowledge among soil decontamination practitioners is related to a preference in favor of conventional methods, like excavation, for the remediation of soil.25 In this article, we explore whether this preference for conventional methods is suggestive of a status quo bias,26 that is, a preference to stick with a previous technological decision even when a superior option is presented. We also explore the strength of this bias by examining the extent to which the preference for conventional methods is maintained even when decision-makers are presented with new information nudging them toward an alternative, more efficient option. To be sure, the decision on which environmental technology to select for a given remediation problem is based on numerous factors, including cost (i.e. whether cheaper alternatives are available), time sensitivity (i.e. how quickly the remediation occurs relative to project imperatives), certainty and liability (i.e. the efficacy of remediation relative to specific targets and objectives) and the broader risk evaluation framework employed (i.e. whether the objective is to manage or eliminate risks from contamination altogether). It is also true that the efficacy of phytoremediation is limited to particular contexts, and that the performance of this method is compromised by such factors as level and depth of contamination, the type of contaminants present, as well as the length of time it takes to decontaminate affected areas so that they reach specified acceptable levels.11 However, a growing body of research has found that phytoremediation works effectively with organic contaminants at relatively low concentrations, and that performance is also effective, – if slower and at times helped by other technologies – for decontaminating trace elements.15,16,18 In cases where phytoremediation offers several advantages over conventional excavation, do practitioners nevertheless show a preference for conventional methods? If so, how deep is this bias for the status quo? 3. Methods and Results To answer these questions, we administered an online survey to the 193 decontamination experts in Quebec, Canada, between November 18 and December 12, 2013. Like other soil decontamination practitioners around the world, these experts possess relevant qualifications in the field of soil remediation, and are accredited with the professional credentials to oversee remediation projects. As a result, these experts also have substantial 5
influence over which remediation technologies are used. Of the 193 experts who received an invitation to complete the survey, 113 (or 59%) filled out the questionnaire partially, while 94 (or 49%) completed it entirely. The survey comprised a short knowledge test on phytoremediation, based on four simple true or false questions: phytoremediation involves plants (true)? Phytoremediation always requires harvesting (false)? Phytoextraction removes organic contaminants (false)? Some plants can extract inorganic pollutants while simultaneously degrading organic contaminants (true)? Any single review published on phytoremediation in the last twenty years contains sufficient information to obtain a perfect score on the test. Yet the mean test score obtained by our experts was 2.18 out of 4 (54%), with a standard deviation of 1.09. This relatively poor score is consistent with the respondents’ self-assessed level of phytoremediation knowledge, with 73% of the sample admitting to having below-average knowledge on this technology. In other words, the survey provides evidence that the large quantity of scientific knowledge produced on phytoremediation since 1995 has failed to reach experts on the ground. Moreover, we find that decontamination professionals working in Quebec show little inclination to recommend phytoremediation, even on parcels of land where scientific research has found this method to be appropriate. For instance, we provided respondents with the description of a site with light and shallow mixed contamination owned by a municipality unwilling to spend much on remediation due to a lack of immediate plans for the area. More importantly, the characterization of the soil provided in the description matched that of soil that has been successfully decontaminated with phytoremediation. 27 We thus loosely based our hypothetical scenario on site characteristics and decontamination profile that made phytoremediation an ideal solution for this particular site. We then asked experts to rate on a 0-6 Likert scale the acceptability of a conventional remediation plan, that is, a plan involving excavation and off-site treatment of the soil. On the same survey vignette, we also asked experts to rate a plan to decontaminate the site with willows and poplars, two of the best-known accumulators. Comparing levels of expert acceptability between the conventional and phytoremediation options, we find that experts are significantly more likely to endorse the former, despite the scientifically documented advantages of phytoremediation in this particular scenario. An un-paired t-test was conducted, revealing a significant difference in the scores for excavation and treatment offsite (M=4.43, SD=1.28) and phytoremediation (M=2.89, SD=1.53); t(220), p=0.00001. Overall, we find that 31% of respondents indicated phytoremediation to be acceptable (i.e. assigned a score of over 3), while only 19% preferred it to costly excavation and off-site treatment (i.e. assigned a higher score to phytoremediation than to excavation). This is true despite the fact that, in this particular instance, phytoremediation presents an easily implementable, low-cost and effective solution for this particular site.
6
To test whether this low acceptability of phytoremediation could be explained by the nature of the site’s contaminants, we presented respondents with a second scenario. In this case, half of the respondents were presented with a site characterized by the light presence of organic substances only, while the other half was assigned a site lightly polluted with trace elements. As in the first scenario, these cases were loosely based on existing research characterizing instances of successful remediation using plants at a given level of contamination. However, relative to trace elements, remediation of organic contaminants with phytoremediation has produced stronger results.11 Respondents exposed to the scenario with organic contaminants might thus be expected to show higher levels of acceptability. Basing our site characterization on this, we found the mean level of acceptance rose to above 3 (on the same 6 point Likert scale) in each case, indicating a higher level of acceptability and perhaps suggesting some acquiescence bias as respondents learned that the survey was on phytoremediation. In any event, we note that this level of acceptability was still far below that observed for excavation and treatment off site (M=4.43, SD=1.28). Moreover, we failed to find a statistically significant difference between mean levels of acceptance across the organic contaminants (M=3.80, SD=1.27) and trace elements (M=3.61, SD 1.50) conditions. We thus find evidence of a status quo bias that appears to systematically work against the uptake of phytoremediation, regardless of the type of contaminants found on site. To make sense of the clear preference for excavation expressed by respondents, we ran a logistic regression and found, after controlling for knowledge and other confounding factors, that the frequency with which experts have recommended excavation in their recent practice decreases the likelihood of preferring phytoremediation. This is exactly what the status quo bias would predict—those having more experience with conventional methods should be more likely to maintain a preference for this method, even when presented with a relatively attractive alternative. As shown in Figure 2, the probability of preferring phytoremediation drops from 0.5 to 0.2 for experts who have recommended excavation more than ten times in recent years. This difference is statistically significant. In other words, accustomed to excavation, experts display a status quo bias, even when controlling for self-assessed knowledge of phytoremediation. Exactly how far does this bias go? Is it resistant to new information regarding the suitability of phytormediation under specific conditions?
7
Figure 2: the effect of recommending excavation more or less frequently on the probability of preferring phytoremediation
Difference significant at p=0.02. To understand the extent to which new scientific research can curb this status quo bias, we embedded an experiment within our survey. We exposed a random subsample of half of the respondents to the abstract of a fictional scientific article presenting unambiguous evidence that phytoremediation using two species of willows will effectively remove Zn, Cu and Pb from polluted soil. To maximize the effect of the treatment, we indicated that the article was from a 2010 issue of Nature and authored by MIT professors. The respondents forming the control group were not exposed to this abstract but, like the members of the experimental group, were asked to re-read the site description presented at the very beginning of the survey (and mentioned above, as the first scenario). Respondents were thus asked to provide a second rating of the phytoremediation plan originally presented in the first scenario they were presented. Crucially, this scenario was characterized by the presence of Zn, Cu and Pb, as well as a remediation plan using willows—just as that featured in the abstract of the fictitious MIT study. Details of this experiment can be found in reference 28. As shown in Figure 3, the subjects who read the article’s abstract increased their acceptability of phytoremediation. The mean acceptability score in the control group is 3.26, compared with 3.81 in the experimental group. This difference proved significant at p=0.05 in a statistical t test. Just to put things into perspective, however, the mean 8
acceptability rate of the conventional plan was 4.4. In other words, exposing on-the-ground experts to scientific knowledge makes a difference, albeit a relatively small one. This reluctance to integrate new information into the assessment of decontamination plans is indicative of a status quo bias that works against the uptake of phytoremediation by decontamination professionals in Quebec. Figure 3: Difference in the acceptability of phytoremediation between control and experimental groups
Difference significant at p=0.05. A careful examination of the reasons given by survey respondents to justify their evaluation of the phytoremediation plan, in conjunction with the justification given for their evaluation of the conventional remediation plan, further supports the status quo bias thesis. In fact, after respondents evaluated each, they were invited to justify their evaluation of each remediation plan. Those who approved phytoremediation wrote comments similar to the following: “Given that time isn’t an issue, this is the most environmentally friendly and least costly of the methods”. This respondent also indicated that excavation and off-site treatment were “too expensive and unnecessary.” Conversely, several respondents who most forcefully rejected phytoremediation justified their evaluation with statements that reveal unreasoned principled objection, an attitude to expect from individuals who are unwilling to consider alternatives to practices to which they are accustomed. One wrote: “this is utopia.” Another simply stated: “this thing doesn’t work.” Meanwhile these same
9
respondents simultaneously affirmed, “excavating the soil is the proper way to go.” Some less vehemently opposed respondents, who nonetheless opposed phytoremediation invoked reasons, most notably time, which made little sense in the context of the specific scenario with which they were presented. Perhaps these respondents had not read the site description carefully enough, but they are perhaps also opposed in principle and simply gave a common generic reason to justify their overly critical evaluation of phytoremediation. Supporting this latter explanation is the fact that several of these respondents generally failed to provide specific reasons to support excavation. Other justifications provide even more straightforward evidence of a status quo bias. One respondent said the method is “not popular enough among fellow experts” to be approved by him. Another one feared that “bankers and public authorities are not familiar enough with the method” to justify advising a client to use phytoremediation. These two respondents also wrote that excavation was acceptable because it is the “classic” way to rehabilitate sites. All of this is consistent with the status quo bias, which is one important barrier to the uptake of new technology in the field of soil remediation. 4. Discussion What do these findings say about the uptake of new environmental technologies in the area of soil decontamination? We find considerable evidence of a status quo bias, working in favor of conventional methods, like excavation and treatment off site, even in cases where phytoremediation presents clear advantages. In fact, we show that respondents most opposed to phytoremediation are those who have recommended conventional off-site treatment methods most frequently, even when controlling for knowledge of phytoremediation. When examined in conjunction with the responses obtained from the open-ended text boxes, our findings suggest that a status quo bias characterizes soil decontamination professionals, tempering the adoption of phytoremediation on sites for which the technology is well suited. While a preference for conventional methods is not evidence in and of itself that such a bias exists, the expression of principled objections— held despite the suitability of phytoremediation to some of the scenarios we presented to practitioners—indicates an evident bias for the status quo. In addition, we find that this status quo bias is relatively “sticky,” and exposure to new scientific knowledge is unlikely to break this decision-making inertia in the uptake of remediation technology. To be sure, our experimental manipulation involved exposing practitioners to the abstract of a single scientific publication. Though we might expect the effect of such limited exposure to be modest, the increase in support for that we do observe does little to close the gap in preferences between phytoremediation and conventional methods. The fact that new information does not increase acceptability to a similar level
10
suggests practitioners are reluctant to update their prior judgments and instead fall back on methods with which they are most familiar. Finally, there are a number of reasons why practitioners might prefer conventional methods to phytoremediation. Beyond a cognitive bias, practitioners might require certainty of outcomes—for a variety of reasons, including liability, for instance—that tilt the calculus in favor of well-known, effective strategies, like excavation. The institutional setting in which practitioners are embedded—with insurers, banks, and even government policy—might also work to make conventional excavation the preferred approach. It is possible that professionals have vested interests in particular technologies, either because excavation services are provided by their firms, or because of institutionalized working relationships that have developed in this sector over the years. While we do not deny that such factors play a role, they are beyond the scope of this particular study. To a certain extent, we also control for some of these factors, by comparing across scenarios that are more and less conducive to phytoremediation. The fact that we find little difference across scenarios, regardless of context, suggests there is an entrenched preference for conventional methods, which we argue is consistent with a status quo bias in the selection of appropriate remediation technology in the Canadian province of Quebec.
5. Conclusion Individuals’ daily behavior is comprised of decisions based on simple cognitive rules and conventions, whether these individuals are experts, politicians or ordinary citizens. 29 Indeed, it would be unreasonable to expect individuals to undertake a careful examination of scientific knowledge every time they had to make a decision, even when they are qualified to read the scientific literature. Individuals are economical and prudent in making decisions. As a consequence, they prefer options closest to their experience over options that require extensive research.30 Such a state of affairs induces a bias that favors the status quo. This cognitive bias encourages and reinforces the type of “lock-in” described by scholars of technology adoption, and which we have observed here in the field of soil decontamination.2 Despite this bias for the status quo in the uptake of new environmental technology, we have shown that—in the case of phytoremediation—scientific knowledge can begin to displace conventions and thereby influence decisions, if only gradually. Scientific knowledge surely does not reach the ground in a straightforward manner (that is, when it does at all), but rather begins to be incorporated into practice through small incremental changes in attitudes. The experimental results discussed here can be viewed as a small but important step in this process, which can improve acceptability at the margin. Overall, they suggest 11
that scientific knowledge can make a difference, and we might expect that the accumulation of scientific research might slowly shift the views of decontamination experts and quite possibly those working in other environmental problem areas as well. Over time, gradual acceptance of new technologies and practices—like phytoremediation—are likely to be influenced by the accumulation of scientific evidence and its communication to practitioners at a level of detail that clearly indicates the benefits, limits and general advantages of the new method relative to those conventionally used. In the case of phytoremediation, this may result in full or partial remediation of large parcels of polluted land currently left abandoned. As the case of phytoremediation suggests, status quo biases stemming from cognitive limitations are likely to continue to pose a barrier to the uptake of new technologies. Awareness of such biases is therefore an important first step in mitigating their impacts. Among practitioners, this means being aware of one’s conventions and being open to learning about alternative methods with some promise. Among scientists, this may imply increasing the effort to disseminate detailed findings in a variety of publication venues. Indeed, our research suggests publication in scientific journals may not be the best way to reach practitioners working on the ground. Success in decontaminating a site with phytoremediation depends on specialized knowledge about the conditions under which the technology functions best. These include the nature of the contaminants, levels of contamination as well as contamination depth. Detailing the strengths and limits of new techniques in a variety of settings is essential when communicating to practitioners who have a variety of criteria in mind when selecting among alternative technologies. Overall, we find that knowledge is key to the successful implementation of a phytoremediation plan on a contaminated site, even among practitioners who are already convinced by the technology’s merits. Reaching professionals on the ground is therefore doubly important: while it encourages acceptance, it also contributes to implementation success.
Acknowledgement The authors acknowledge the financial support of Genome Quebec, Genome Canada and the Social Sciences and Humanities Research Council of Canada. They also thank the members of Genorem for their contribution to this research, as well as Monika Smaz and Irena Nedeva who provided excellent research assistance.
References 1.
Roupas, P. Human and Organisational Factors Affecting Technology Uptake by 12
Industry. Innov. Manag. Policy Pract. 10, 4–28 (2008). 2. Arthur, W. B. Positive Feedbacks in the Economv. Sci. Am. 262, 92–99 (1990). 3. Panagos, P., Van Liedekerke, M., Yigini, Y. & Montanarella, L. Contaminated Sites in Europe: Review of the Current Situation Based on Data Collected through a European Network. J. Environ. Public Health 2013, 1–11 (2013). 4. Search for Superfund Sites Where You Live | Superfund | US EPA. Available at: http://www.epa.gov/superfund/search-superfund-sites-where-you-live. (Accessed: 22nd February 2016) 5. McIntyre, T. in Phytoremediation (ed. Tsao, D.) 78, 97–123 (Springer Berlin Heidelberg, 2003). 6. Johnson, B. L. & DeRosa, C. The Toxicologic Hazard of Superfund Hazardous-Waste Sites. Rev. Environ. Health 12, (1997). 7. Chou, C. L., Paon, L. A. & Moffatt, J. D. Cadmium, Copper, Manganese, Silver, and Zinc in Rock Crab ( Cancer irroratus ) from Highly Copper Contaminated Sites in the Inner Bay of Fundy, Atlantic Canada. Bull. Environ. Contam. Toxicol. 68, 885–892 (2002). 8. Barrento, S. et al. Essential Elements and Contaminants in Edible Tissues of European and American Lobsters. Food Chem. 111, 862–867 (2008). 9. Ali, H., Khan, E. & Sajad, M. A. Phytoremediation of Heavy Metals: Concepts and Applications. Chemosphere 91, 869–881 (2013). 10. Salt, D. E., Smith, R. D. & Raskin, I. Phytoremediation. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 643–668 (1998). 11. Batty, L. C. & Dolan, C. The Potential Use of Phytoremediation for Sites With Mixed Organic and Inorganic Contamination. Crit. Rev. Environ. Sci. Technol. 43, 217–259 (2013). 12. Thewys, T., Witters, N., Meers, E. & Vangronsveld, J. Economic Viability of Phytoremediation of a Cadmium Contaminated Agricultural Area Using Energy Maize. Part II: Economics of Anaerobic Digestion of Metal Contaminated Maize in Belgium. Int. J. Phytoremediation 12, 663–679 (2010). 13. Lewandowski, I., Schmidt, U., Londo, M. & Faaij, A. The Economic Value of the Phytoremediation Function – Assessed by the Example of Cadmium Remediation by Willow (Salix ssp). Agric. Syst. 89, 68–89 (2006). 14. Salt, D. E. et al. Phytoremediation: A Novel Strategy for the Removal of Toxic Metals from the Environment Using Plants. Nat. Biotechnol. 13, 468–474 (1995). 15. Cameselle, C., Chirakkara, R. A. & Reddy, K. R. Electrokinetic-Enhanced Phytoremediation of Soils: Status and Opportunities. Chemosphere 93, 626–636 (2013). 16. Leung, H.-M. et al. Interactions Between Arbuscular Mycorrhizae and Plants in Phytoremediation of Metal-Contaminated Soils: A Review. Pedosphere 23, 549–563 (2013). 17. Lee, J. H. An Overview of Phytoremediation As a Potentially Promising Technology for Environmental Pollution Control. Biotechnol. Bioprocess Eng. 18, 431–439 (2013). 18. Gupta, D. K., Huang, H. G. & Corpas, F. J. Lead Tolerance in Plants: Strategies for Phytoremediation. Environ. Sci. Pollut. Res. 20, 2150–2161 (2013). 19. Lokhande, V. H., Gor, B. K., Desai, N. S., Nikam, T. D. & Suprasanna, P. Sesuvium Portulacastrum: a Plant for Drought, Salt Stress, Sand Fixation, Food and Phytoremediation. A Review. Agron. Sustain. Dev. 33, 329–348 (2013). 20. Agostini, E., Talano, M. A., González, P. S., Oller, A. L. W. & Medina, M. I. Application of Hairy Roots for Phytoremediation: What Makes them an Interesting Tool for this Purpose? Appl. Microbiol. Biotechnol. 97, 1017–1030 (2013). 13
21. Kanu, S. A., Okonkwo, J. O. & Dakora, F. D. Aspalathus Linearis (Rooibos tea) as Potential Phytoremediation Agent: a Review on Tolerance Mechanisms for Aluminum Uptake. Environ. Rev. 21, 85–92 (2013). 22. Tak, H. I., Ahmad, F. & Babalola, O. O. in Reviews of Environmental Contamination and Toxicology Volume 223 (ed. Whitacre, D. M.) 223, 33–52 (Springer New York, 2013). 23. Rathod, P. H., Rossiter, D. G., Noomen, M. F. & van der Meer, F. D. Proximal Spectral Sensing to Monitor Phytoremediation of Metal-Contaminated Soils. Int. J. Phytoremediation 15, 405–426 (2013). 24. Lachapelle, E. & Montpetit, E. Putting Phytoremediation into Action. The Scientist 29, 18–19 (2015). 25. Montpetit, É. & Lachapelle, E. Information, Values and Expert Decision-Making: the Case of Soil Decontamination. Policy Sci. (2015). doi:10.1007/s11077-015-9225-x 26. Samuelson, W. & Zeckhauser, R. Status Quo Bias in Decision Making. J. Risk Uncertain. 1, 7–59 (1988). 27. Doyle, C. La Phytoremédiation: une solution à la contamination des sites de traitement du bois? (Université de Sherbrooke, 2008). 28. Montpetit, E. & Lachapelle, E. Can Policy Actors Learn from Academic Scientists? Environ. Polit. 24, 661–680 (2015). 29. Kahneman, D. Maps of Bounded Rationality: Psychology for Behavioral Economics. Am. Econ. Rev. 93, 1449–1475 (2003). 30. Lindblom, C. E. The Science of ‘Muddling Through’. Public Adm. Rev. 19, 79–88 (1959).
14