SEARCH

SEARCH BY CITATION

Keywords:

  • best-worst scaling;
  • biodiversity loss;
  • conservation policy;
  • conservation strategies;
  • latent class;
  • triage
  • clase latente;
  • escala mejor-peor;
  • estrategias de conservación;
  • pérdida de biodiversidad;
  • políticas de conservación;
  • triaje

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

Abstract: The large investments needed if loss of biological diversity is to be stemmed will likely lead to increased public and political scrutiny of conservation strategies and the science underlying them. It is therefore crucial to understand the degree of consensus or divergence among scientists on core scientific perceptions and strategies most likely to achieve given objectives. I developed an internet survey designed to elucidate the opinions of conservation scientists. Conservation scientists (n =583) were unanimous (99.5%) in their view that a serious loss of biological diversity is likely, very likely, or virtually certain. Scientists’ agreement that serious loss is very likely or virtually certain ranged from 72.8% for Western Europe to 90.9% for Southeast Asia. Tropical coral ecosystems were perceived as the most seriously affected by loss of biological diversity; 88.0% of respondents familiar with that ecosystem type agreed that a serious loss is very likely or virtually certain. With regard to conservation strategies, scientists most often viewed understanding how people and nature interact in certain contexts and the role of biological diversity in maintaining ecosystem function as their priorities. Protection of biological diversity for its cultural and spiritual values and because of its usefulness to humans were low priorities, which suggests that many scientists do not fully support the utilitarian concept of ecosystem services. Many scientists expressed a willingness to consider conservation triage, engage in active conservation interventions, and consider reframing conservation goals and measures of success for conservation of biological diversity in an era of climate change. Although some heterogeneity of opinion is evident, results of the survey show a clear consensus within the scientific community on core issues of the extent and geographic scope of loss of biological diversity and on elements that may contribute to successful conservation strategies in the future.

Resumen: Las grandes inversiones requeridas para frenar la pérdida de diversidad biológica probablemente llevarán a un mayor escrutinio público y político de las estrategias de conservación y de la ciencia que las sustenta. Por lo tanto, es crucial entender el grado de consenso o divergencia entre científicos sobre las percepciones y estrategias científicas planteadas para alcanzar ciertos objetivos. Desarrollé un muestreo por internet para aclarar las opiniones de científicos de la conservación. Científicos de la conservación (n =583) fueron unánimes (99.5%) en opinar que una pérdida severa de diversidad biológica es posible, muy posible o virtualmente cierta. El acuerdo de los científicos respecto a que una pérdida severa es muy probable o virtualmente cierta varió de 72.8% en Europa Occidental a 90.9 en el sureste de Asia. Los ecosistemas de corales tropicales fueron percibidos como los más afectados por la pérdida de diversidad biológica; 88.0% de los encuestados familiarizados con ese tipo de ecosistema estuvo de acuerdo en que una pérdida severa es muy probable o virtualmente cierta. En relación con las estrategias de conservación, los científicos a menudo visualizaron como prioritario el entendimiento la interacción entre la gente y la naturaleza en ciertos contextos y el papel de la diversidad biológica en el mantenimiento de las funciones del ecosistema. La protección de la diversidad biológica por sus valores culturales y espirituales y por su utilidad para los humanos fueron prioridades bajas, lo que sugiere que muchos científicos no apoyan el concepto utilitario de los servicios ecosistémicos. Muchos científicos expresaron disponibilidad para considerar el triaje de conservación, involucrarse activamente en intervenciones de conservación y considerar el replanteamiento de metas de conservación y medidas de éxito para la conservación de la diversidad biológica en una era de cambio climático. Mientras que la heterogeneidad de opiniones es evidente, los resultados del muestreo muestran un consenso claro entre la comunidad científica respecto a temas centrales de la extensión y alcance geográfico de la pérdida de diversidad biológica y sobre elementos que pueden contribuir a estrategias de conservación exitosas en el futuro.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

The widespread loss of biological diversity poses a challenge for countries around the world for ecological, economic, and social reasons (Millennium Ecosystem Assessment 2005; SCBD 2010; TEEB 2010). As the extent of loss of biological diversity and its potential effects has gained recognition, a variety of policy responses have followed. For example, numerous national and state-level jurisdictions have implemented policy and regulations to protect endangered species, the International Union for Conservation of Nature Red List and Convention on Biological Diversity were created, and the Intergovernmental Panel on Biodiversity and Ecosystem Services (IPBES) was formed recently. Responding to loss of biological diversity will require improvements in scientific modeling, long-term monitoring, active restoration efforts, and the creation and refinement of governance institutions (e.g., Pereira et al. 2010; Jones et al. 2011; Perrings et al. 2011), all of which will require significant investment.

The investments needed if loss of biological diversity is to be stemmed will likely lead to increased public and political scrutiny of conservation strategies and the science underlying them. As in the climate-change debate (Anderegg et al. 2010; Rosenberg et al. 2010), supporters of strong action will likely argue that the science is clear and overwhelmingly consistent with action, whereas skeptics will argue that the science is highly uncertain and unworthy of use as a foundation for public policy. Thus, it will be crucial to understand the degree of consensus among conservation professionals on core scientific points and on feasible and preferred management interventions.

Through an internet survey, I assessed the degree of convergence of scientific understanding and opinions on conservation strategies among conservation scientists. The survey addressed three themes: geographic and temporal scope of loss of biological diversity and scientists’ level of understanding of the causes and effects of this loss; perceived importance of different conservation values; and level of agreement on some potentially controversial views in conservation science. The results from my survey provide a snapshot of scientists’ current opinions on the loss of biological diversity globally and conservation strategies that might be used to reverse that loss. Understanding scientists’ opinions could help policy makers interpret conflicting scientific advice, identify conservation options, and improve the likelihood of successful implementation of conservation initiatives.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

Survey Instrument

The survey had six sections. The first introduced the research and addressed survey ethics and data security (see Supporting Information for an example of a full survey). The second section asked about respondents’ demographic and professional characteristics. The third section asked respondents’ opinions about the geographic and temporal scope of loss of biological diversity by region and ecosystem type (“major habitat types” of Olson and Dinerstein [1998]). I modified several questions from a survey that examined the views of climate scientists on the nature, causes, and effects of climate change (Rosenberg et al. 2010) and used language recommended for climate-science research (Morgan et al. 2009) to describe probabilities in qualitative and quantitative terms. The fourth section included a set of 16 best–worst scaling (BWS) questions (Flynn et al. 2007) to fully rank respondents’ levels of agreement with 32 value-oriented statements developed by Sandbrook et al. (2011). The statements encompassed four dimensions within which diverse opinions had been expressed by young conservation scientists: values (why people care); priorities (how they should be set); geography (where conservation should take place); and actions (how conservation should be undertaken). The fifth section of the survey asked about respondents’ levels of agreement with potentially controversial statements about conservation strategies. Of the 17 statements presented in the fifth section, I based 14 on responses from conservation experts interviewed by Hagerman et al (2010), who interviewed a small group of experts on biological diversity and adaptation to climate change. I emphasized conservation triage (i.e., the explicit decision not to treat a given population, species, or ecosystem knowing that a lack of effort will likely lead to possible extinction of that population or species or extreme alteration of the ecosystem). A final section, not reported here, queried respondents regarding possible reasons for a lack of conservation science uptake by policy makers. The survey was constructed with Sawtooth Software's (Sequim, Washington) SSI web-based interviewing platform.

Sample Frame and Sample

My sample frame included all authors who published from 2005 to 2010 in 19 international journals with downloadable abstracts and author information: AMBIO; Animal Conservation; Aquatic Conservation: Biodiversity and Conservation; Biological Invasions; BioScience; Conservation Biology; Conservation Letters; Diversity and Distributions; Ecological Economics; Environmental Conservation; Fish and Fisheries; Frontiers in Ecology and the Environment; Global Ecology and Biogeography; Human Dimensions of Wildlife; Insect Conservation and Diversity; Journal of Applied Ecology; Marine and Freshwater Ecosystems; Restoration Ecology; and Society and Natural Resources.

I downloaded abstracts for 10,972 articles to Endnote. From this pool, I selected 531 articles (4.8% of the total), which included research articles, reviews, and editorials. The number of articles in the sample was proportional to the total number of articles published annually in each journal (Supporting Information). For journals that publish research not directly pertinent to the management of biological diversity (e.g., Ecological Economics publishes articles on energy policy and trade), I drew proportionally fewer articles overall, selecting only those with a clear biological-diversity theme. From these articles I drew a sample of 1826 authors to contact. I used email addresses provided in the articles or found them through a Google search. For each respondent that completed the survey, I calculated Hirsh's h index (Hirsch 2005), a proxy for scientific standing, with the bibliometric software Publish or Perish (version 3.1; Harzing 2010). An author who has published 50 papers with at least 50 citations each, for example, would have an h index of 50. Although h index is subject to potential errors (e.g., not all publications are found), it provides a relatively robust means of calculating scientific standing. For analysis, I aggregated h index scores into five categories with roughly equal numbers of respondents: ≤3 (n= 124); 4–6 (n= 103); 7–11 (n= 121); 12–19 (n= 123); and ≥20 (n= 112).

Survey Implementation

I contacted authors up to five times (Dillman et al. 2009) over five weeks. I emailed a notice on 19 February 2011 that the survey was to follow and the main survey invitation on 1 March 2011. The invitation email contained a web link so that respondents could access the survey with a single click. I also sent the following additional emails (Supporting Information): a reminder notice on 6 March 2011 to nonrespondents, a second survey invitation to nonrespondents on 13 March 2011, and a final reminder to remaining nonrespondents on 29 March 2011.

Data Analyses

I used Sawtooth Software's BWS module to generate 300 versions of the BWS questions (Supporting Information) to which respondents were randomly assigned. Each of 16 BWS questions contained four statements about conservation strategies, scope, and values drawn from the full set of 32 statements. Respondents were asked to choose the statements with which they agreed most and least (Supporting Information). I analyzed data from the BWS ranking exercise with Sawtooth Software's Hierarchical Bayesian procedures (Sawtooth Software 2007).

I used respondents’ assessments of the likelihood of climate change and serious losses in biological diversity and their ratings of scientists’ level of understanding of the causes and consequences of losses of biological diversity in 2-stage latent-class (LC) cluster analyses. First, LC clustering identified indicator variables that were strongly associated with an unobservable latent variable. Second, segmentation analyses identified mutually exclusive demographically or professionally based segments that were predictive of posterior class membership probabilities from the LC models.

In the first-stage LC model, the frequency of particular responses was used to estimate model parameters with the expectation-maximization algorithm (Vermunt & Magidson 2002) such that the expected frequencies were as close as possible to observed frequencies (see Eid et al. [2003] for an overview of the methods). Heterogeneity of responses can be explored with a variety of tests and information criteria that identify the number of subsegments that result in the closest matches between expected and observed response patterns. Because there is no definitive test of best fit, it is common to use information criteria, which impose penalties on extra model parameters, to choose final models. I used Bayesian information criteria (BIC) to identify the model that was most parsimonious. The BIC test is more appropriate for relatively simple LC models and favors models with fewer classes than alternative criteria (e.g., Akaike information criteria), which facilitates model interpretation.

LC models assume local independence between indicator variables. That is, responses on items are independent given class membership (Eid et al. 2003), which implies that LC structure explains all associations between items that are based on observable responses. I tested the local independence assumption with the bivariate residual Pearson χ2 statistic. A significant bivariate residual statistic (i.e., χ2 > 3.84, df = 1, p < 0.05) indicates the assumption of local independence is erroneous. When I found significant interactions between indicators, I sequentially deleted indicators with the highest number of significant bivariate residual statistics until all significant interactions were eliminated. Indicator variables not included in the LC cluster analysis were not needed to differentiate heterogeneity within the sample. I used Latent Gold software (Vermunt & Magidson 2005) to estimate all LC cluster models.

In the second-stage model, I used chi-squared goodness-of-fit tests to identify significant predictors of LC membership patterns and merge predictor categories that did not differ in their prediction of the dependent variables (Magidson & Vermunt 2005). Covariates I tested as potential predictors were age; gender; region of residence; sector (academic or other); discipline (biological sciences or other); h index; and journal from which the respondent was drawn in the sample. I used the Chi-Squared Automatic Interaction Detection (CHAID) software (Magidson 2005) to systematically test all possible combinations of predictor variables and identify all those that were statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

Understanding of Loss of Biological Diversity

Respondents were unanimous (99.5%) in their view that it is likely a serious loss of biological diversity is underway at a global extent; 8.4%, 24.9%, and 66.2% thought that serious loss is likely, very likely, or virtually certain, respectively (Table 1). There was even greater consensus (79.1%) that human activities are virtually certainly accelerating the loss of biological diversity. By comparison, 61.9% and 55.1% thought climate change (presented as “global warming” in the survey to match language used by Rosenberg et al. [2010]) is a process that is already underway and that humans are accelerating it, respectively. This is consistent with results from Rosenberg et al. (2010), who found that in 2005, 61.6% and 49.2% of U.S. climate scientists strongly agreed that climate change was already underway and that human activities were accelerating climate change, respectively. The majority of respondents agreed (58.1%) or strongly agreed (14.6%) that the nature and causes of loss of biological diversity are highly understood (Supporting Information). Scientists agreed (35.5%) or strongly agreed (6.2%) that the consequences of the loss of biological diversity are highly understood.

Table 1.  Respondents’ (n= 583) estimates of probabilities that loss of biological diversity and climate change are occurring and being accelerated by human activities.a
Statement about biological diversity or climate changeVirtually impossibleVery unlikelyUnlikelyAbout an even chanceLikelyVery likelyVirtually certain
  1. aLanguage recommended by the U.S. Climate Change Science Program (Morgan et al. 2009): virtually impossible, ≤0.01 probability; very unlikely, approximately 0.01–0.20 probability; unlikely, less than even chance (i.e., 0.20–0.50 probability); about an even chance, approximately 0.50 ± 0.05 probability; likely, very likely, and virtually certain, 0.50–0.80, 0.80–0.99, and ≥0.99 probabilities, respectively.

  2. bStatement was retained as a covariate in a latent-class cluster analysis of scientific understanding of loss of biological diversity.

A serious loss of biological diversity is underway at the global scaleb0.0000.0000.0030.0020.0840.2490.662
Human activities are accelerating the loss of biological diversity at the global scale0.0000.0000.0000.0030.0330.1730.791
Global warming is a process that is already underwayb0.0000.0020.0030.0260.0770.2730.619
Human activities are accelerating global warmingb0.0000.0020.0020.0120.1150.3190.551

The 5-class LC cluster model minimized BIC and was chosen for further refinement. One bivariate residual was significant at the 5% level, indicating some redundancy between the 2 climate-change indicators. Dropping one statement (“Global warming is a process that is already underway.”) from the model eliminated the significant bivariate residual. The final model (n= 582, 44 parameters, entropy R2= 0.78, classification error = 10.2%) cleaved the sample into five distinct clusters (Fig. 1): alarmed, concerned, science optimists, moderates, and science pessimists. All LC cluster analyses were conducted with data from only 582 of 583 respondents because I dropped one invalid response.

image

Figure 1. Membership in latent-class clusters (y-axis) on the basis of scientists’ understanding of loss of biological diversity (1, virtually impossible; 2, very unlikely; 3, unlikely; 4, about an even chance; 5, likely; 6, very likely; 7, virtually certain; SD, strongly disagree; D, disagree; N, neither agree nor disagree; A, agree; and SA, strongly agree; complete statements: indicator 1, A serious loss of biological diversity is underway at the global scale. 2, Human activities are accelerating the loss of biological diversity at the global scale. 3, Human activities are accelerating global warming. 4, Scientists have a strong understanding of the nature and causes of changes in biological diversity. 5, Scientists have a strong understanding of the consequences of changes in biological diversity.).

Download figure to PowerPoint

Cluster 1, alarmed, contained 60.8% of the sample. Respondents in this cluster were very (8.9%) or virtually (91.9%) certain that a serious loss of biological diversity is underway and every respondent believed human activities are accelerating the loss. Those in cluster 2 (22.4% of the sample), concerned, expressed similar views as those in the alarmed cluster, but they were more measured in their views of the level of seriousness of biological diversity loss and human activities as drivers of that loss (and climate change). LC clusters 3 (7.0%, science optimists) and 5 (4.2%, science pessimists) held very similar views to each other on the seriousness of biological diversity loss, but they differed greatly on their views of scientists’ understanding of the causes and effects of that loss. Cluster 4 (5.6%, moderates) respondents were more measured in all their responses.

In the subsequent CHAID analysis, only publication in Conservation Biology explained significant variation (χ2= 10.94, df = 4, p= 0.03) in probability of membership in LC clusters (Supporting Information). Respondents who had published in Conservation Biology were more likely to belong to alarmed, science optimists, and science pessimists clusters. The common theme among these three clusters was that 100% of respondents viewed a serious loss of biological diversity as very likely or virtually certain.

Geographic and Temporal Scope of Loss of Biological Diversity

I asked respondents to provide only answers for regions and major ecosystem types with which they were familiar. I interpreted no response as do not know. For regions or ecosystems where respondents thought loss of biological diversity was very likely or virtually certain, I asked a follow-up question regarding the timing of loss. Respondents’ agreement that serious biological diversity loss was very likely or virtually certain ranged from lows of 72.8% (27.0% very likely, 45.8% virtually certain) for Western Europe to highs of 90.9% (33.0% very likely, 57.9% virtually certain) for Southeast Asia (Supporting Information).

The ecosystem respondents viewed as most seriously affected by loss of biological diversity was marine tropical coral; 38.7% and 49.3% believed that a serious loss of biological diversity in marine tropical coral is very likely or virtually certain, respectively (Supporting Information). Tropical moist and dry broadleaf forest and mangrove ecosystems were also viewed as subject to serious levels of loss (loss virtually certain = 47.4%, 44.6%, and 40.6%, respectively), whereas serious losses of biological diversity in marine upwelling ecosystems was viewed as virtually certain by 17.9% of respondents.

Opinions on the timing of the most serious losses in biological diversity over and within the range of ecosystems were broad (Supporting Information). Generally, respondents thought serious losses of biological diversity in freshwater and temperate terrestrial ecosystems tended to occur more in the past relative to tropical and polar ecosystems.

Conservation Values and Priorities

All 583 respondents completed 16 BWS ranking questions (Table 2). Mean scores, which represent likelihood of being chosen as the statement that respondents most agreed with, summed to 100. An item with a mean score of 6 was thus twice as likely to be chosen as that most agreed with by respondents as an item with a mean score of 3. The distribution of mean scores exhibited some discontinuity. The two statements with the highest rank had significantly higher levels of agreement (95% CI of 6.824–7.084 for “Conservation planning needs to understand how people and nature interact in particular places.” and 6.275–6.633 for “Biological diversity should be conserved because it sustains ecosystem function.”) than the statement ranked third (95% CI of 5.041–5.434 for “Conservation priorities should reflect the need to protect globally important species and ecosystems.”). Respondents were very unlikely to select the two statements with the lowest rank as ones they most agreed with (95% CI of 0.543–0.765 for “The value of biological diversity depends on its usefulness to people.” and of 0.371–0.483 for “Long-term residents should be displaced from protected areas if conservation needs warrant.” Other statements filled the gradient between the extremes (Supporting Information).

Table 2.  Statements of conservation values, priorities, strategies, or actions that respondents agreed with most and least.
Statement and overall rank of statementTimes statement shown to respondentsTimes selected as most agreed with (%)Times selected as least agreed with (%)Likelihood of selection as most agreed with (%), 95% CI%
1. Conservation planning needs to understand how people and nature interact in particular places.1166649 (55.7)49 (4.2)6.824–7.084
2. Biological diversity should be conserved because it sustains ecosystem function.1166607 (52.1)75 (6.4)6.275–6.633
3. Conservation priorities should reflect the need to protect globally important species and ecosystems.1168483 (41.4)122 (10.4)5.041–5.434
4. Conservation success demands significant changes in human population growth.1162514 (44.2)168 (14.5)4.880–5.407
5. People should be offered incentives to change their behavior to conserve species and ecosystems.1166465 (39.9)139 (11.9)4.863–5.280
6. Conservation should prevent the human-caused extinction of species.1170447 (38.2)128 (10.9)4.808–5.193
7. The best way to understand what works in conservation is through the systematic comparative analysis of multiple cases or experiments.1164435 (37.4)141 (12.1)4.586–4.963
8. Conservation efforts should also address poverty alleviation.1170384 (32.8)204 (17.4)3.882–4.366
9. Science should be used to determine—not simply inform—policy and management decisions affecting biological diversity.1156395 (34.2)277 (24.0)3.806–4.341
10. Humans have a moral duty to conserve biological diversity.1169377 (32.2)209 (17.9)3.721–4.183
11. People should be made to change their behavior to conserve species and ecosystems.1169362 (31.0)221 (18.9)3.618–4.055
12. Conservation success demands dramatic changes in life-styles of the world's rich.1167363 (31.1)256 (21.9)3.541–4.042
13. Conservation planning should concentrate on key priorities, instead of spreading effort across all locations.1167311 (26.6)184 (15.8)3.303–3.685
14. Conservation effort should be focused on creating protected areas of high biological diversity.1173306 (26.1)207 (17.6)3.261–3.661
15. To be effective, conservation planning must be done locally.1170312 (26.7)208 (17.8)3.251–3.651
16. All species have a right to exist.1165275 (23.6)275 (23.6)3.001–3.454
17. Successful conservation demands the strict enforcement of regulations and laws.1169291 (24.9)257 (22.0)2.970–3.406
18. Biological diversity should be conserved because of its potential future values.1166275 (23.6)255 (21.9)2.692–3.059
19. Conservation action should be focused on areas where it can be most cost-effective.1164256 (22.0)312 (26.8)2.418–2.811
20. Conservation action is needed in areas extensively modified by human activity.1172238 (20.3)286 (24.4)2.353–2.708
21. Conservation success demands the de-carbonization of the global economy.1167217 (18.6)318 (27.2)2.253–2.643
22. There should be conservation areas free from any human influence.1162210 (18.1)358 (30.8)2.017–2.390
23. Biological diversity should be conserved to ensure human survival.1162224 (19.3)381 (32.8)2.008–2.411
24. The best way to understand what works in conservation is the in-depth study of individual cases.1163169 (14.5)292 (25.1)1.798–2.087
25. Effective conservation planning must be based on geographic information science.1177174 (14.8)347 (29.5)1.617–1.901
26. Conservation priorities should be set by the people most affected by them.1166122 (10.5)435 (37.3)1.195–1.484
27. Biological diversity should be conserved because of its cultural and spiritual value.1161106 (9.1)359 (30.9)1.149–1.375
28. Trade in wild species and their products can work as a tool for conservation.117189 (7.6)522 (44.6)0.813–1.038
29. Conservation must do no harm to human communities.115880 (6.9)517 (44.6)0.730–0.933
30. Biological diversity should be conserved because of the beauty of nature.116469 (5.9)530 (45.5)0.654–0.834
31. The value of biological diversity depends on its usefulness to people.115788 (7.6)672 (58.1)0.543–0.765
32. Long-term residents should be displaced from protected areas if conservation needs warrant.116535 (3.0)624 (53.6)0.371–0.483

Management and Policy Opinions

The 17 statements about conservation interventions, triage, and the management of biological diversity were used in a LC analysis of respondents’ conservation-strategy orientation. A 6-class LC cluster model minimized BIC. Twenty-one bivariate residuals were significant at the 5% level in the first model, indicating substantial local dependence among the conservation-strategy indicators. I sequentially deleted eight indicators (statements 3.2, 3.2, 1.4, 3.6, 3.1, 2.2, 2.4, and 3.4) to eliminate all significant bivariate residuals. The final model (n= 582 respondents, 86 parameters) was well supported by the data (entropy R2= 0.73, classification error = 14.3%). The classes (Fig. 2) were indicative of the diversity of opinions held by scientists on strategic approaches to conservation. The indicator variables dropped from the analysis are fully described in Supporting Information.

image

Figure 2. Membership in latent-class clusters (y-axis) on the basis of scientists’ level of agreement with potentially controversial conservation management strategies (cluster 6, protesters [n =3], not included) (SD, strongly disagree; D, disagree; N, neither agree nor disagree; A, agree; SA, strongly agree; full statements with which respondents were presented: 1.1, “We should be helping species adapt by letting them stay natural and letting the processes go as they will as climate changes.” 1.2, “Assisted migration interventions are doomed to failure. Our history of biological manipulation has not gone well and there is no reason to think that future manipulations will go better.” 1.3, “Climate change is going to force our hand. We need to use assisted migration to move species that can't get around urban and agricultural barriers to places where they are going to be more likely to persist.” 1.5, “We don't have the framework for tolerating loss. We have to figure out, for critical ecosystems to start with, what are the minimum number of species within functional groups that are essential for ecosystem services? We need to protect them even if we lose others.” 2.1, “Inevitably one has to make some harsh decisions such as what you give up on. No doubt there will be species that we should and will give up on.” 2.3, “We have spent tons of money trying to save some icon species. If we went purely from a triage perspective, we would have let those species go extinct. But if an icon species can attract extra money for conservation, it is not taking resources from other conservation programs. Triage could thus harm conservation efforts by limiting our capacity to raise money.” 2.5, “We cannot justify major triage choices because we don't know the role of particular species in ecosystems.” 3.5, “We need more rules, better monitoring, increased enforcement, and larger fines. Making damaging human behavior illegal and expensive is central to any strategy meant to protect biological diversity.” 3.7, “Conserving biological diversity in an era of climate change means conservation professionals need to be willing to rethink conservation goals and standards of success.”).

Download figure to PowerPoint

One cluster, mainstream moderates, composed 43.2% of the sample; respondents in this cluster tended to be relatively neutral on most statements. The distinguishing characteristics of a second naturally oriented cluster (32.0% of the sample) was respondents’ relatively strong focus on helping species stay natural (indicator 1.1 [“We should be helping species adapt by letting them stay natural and letting the processes go as they will as climate changes.”]), pessimism regarding assisted migration (indicators 1.3 [“Climate change is going to force our hand. We need to use assisted migration to move species that can't get around urban and agricultural barriers to places where they are going to be more likely to persist.”] and 1.5 [“We don't have the framework for tolerating loss. We have to figure out, for critical ecosystems to start with, what are the minimum number of species within functional groups that are essential for ecosystem services? We need to protect them even if we lose others.”]), and unwillingness to protect some species at the expense of others (indicator 2.1 [“Inevitably one has to make some harsh decisions such as what you give up on. No doubt there will be species that we should and will give up on.”]). Respondents in this cluster were relatively neutral on triage issues.

Respondents in the third cluster (13.2% of the sample), interventionists, were more supportive of direct conservation interventions. They strongly agreed that conservation goals needed rethinking (indicator 3.7 [“Conserving biological diversity in an era of climate change means that conservation professionals need to be willing to rethink conservation goals and standards of success.”]), that the use of triage should not be discounted because of a lack of ecological knowledge (indicator 2.5 [“We cannot justify major triage choices because we don't know the role of particular species in ecosystems.”]), and that conservation actions should not be taken for some species (indicator 2.1). Although they were supportive of assisted migration, they also expressed some pessimism about the probability of success of this strategy (indicators 1.2 [“Assisted migration interventions are doomed to failure. Our history of biological manipulation has not gone well and there is no reason to think that future manipulations will go better.”] and 1.5).

Respondents in cluster 4, preservationists, did not agree with assisted migration (indicator 1.3), disagreed that some species should be protected at the expense of others (indicators 1.5 and 2.1), and were supportive of more regulation of human behavior (indicator 3.5 [“We need more rules, better monitoring, increased enforcement, and larger fines. Making damaging human behavior illegal and expensive is central to any strategy meant to protect biological diversity.”]).

Respondents in cluster 5 (4.0% of the sample), conservationists, did not agree that some species should be protected at the expense of others (indicators 1.5 and 2.1), were supportive of assisted migration efforts (indicators 1.2 and 1.3), skeptical of triage (indicators 2.3 [“We have spent tons of money trying to save some icon species. If we went purely from a triage perspective, we would have let those species go extinct. But if an icon species can attract extra money for conservation, it is not taking resources from other conservation programs. Triage could thus harm conservation efforts by limiting our capacity to raise money.”] and 2.5), and were quite supportive of more conservation rules and their enforcement (indicator 3.5). Only three respondents were in cluster 6, protestors. They strongly disagreed with most statements and either did not pay attention to survey questions or were exhibiting protest responses because they did not like the questions. Their responses are not included in Fig. 2.

Opinions regarding effective conservation strategies differed significantly (χ2= 23.25, df = 5, p= 0.01) among residents of the following 2 groups of countries or regions: (1) Africa, Asia, and Europe and (2) Australia, New Zealand, Pacific Islands, North America, Latin America, and Caribbean. Membership in the interventionist cluster was over 10% higher for scientists from the second group, whereas scientists from the first group were 6.8% more likely to be members of the preservationist cluster (Supporting Information). No other significant differences in the predictive ability of any of the professional or demographic covariates in the model were detected. Alternatively, opinions regarding conservation strategies differed significantly on the basis of h index (χ2= 16.29, df = 5, p= 0.09). Scientists with h≥ 13 were almost 12% less likely to be members of the naturally-oriented cluster and almost 9% more likely to be members of the interventionist cluster than scientists with h < 13 (Supporting Information).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

Understanding of Loss of Biological Diversity

There was striking agreement among scientists on the overall extent and geographic scope of the loss of biological diversity but a lower degree of consensus on the timing of the most serious losses in different regions. Rosenberg et al. (2010) found a similar pattern among climate scientists, who agreed on core issues of the nature and causes of climate change regardless of professional or demographic characteristics but agreed less on timelines for significant impacts in specific regions. The degree of convergence in expert opinion can be important to policy makers because it can provide important information about ecological certainties and uncertainties, and the likely outcome of different policy options (Lubchenco 1998; Anderegg et al. 2010). Convergence of expert opinion may help eliminate policy options that are not supported by scientific consensus (Rudd 2011).

I identified five distinct clusters of scientists for whom opinions on scientific understanding regarding biological diversity loss varied significantly. The only covariate that was significantly associated with cluster membership was whether the survey respondent had published an article in Conservation Biology. This journal may draw more authors that view global losses of biological diversity as a major societal challenge. Survey results showed that scientists’ opinions regarding the loss of biological diversity could not be predicted on the basis of any other observable demographic or professional characteristics. This suggests a high overall degree of cohesiveness of core values within clusters of scientists.

Conservation Values and Priorities

Sandbrook et al. (2011) found that young conservation scientists (n= 64) hold a plurality of values and opinions. My survey sampled a much broader range of conservation professionals at all career levels. Respondents, in aggregate, placed a high level of importance on context-dependent understanding of how people and nature interact. They also ranked the “role of biological diversity in maintaining ecosystem function” highly. This phrase, used by Sandbrook et al. (2011), was likely interpreted in the context of species and populations by respondents in my survey. Other high-ranking items focused on goals (protecting globally important species and ecosystems, preventing human-caused extinction), methods (comparative analysis of multiple cases or experiments), and social factors (limiting human population growth and using incentives to alter behavior). Respondents placed less emphasis on protecting biological diversity because of its cultural or spiritual values or because of its usefulness to humans.

Sandbrook et al. (2011) identified four distinct value-based segments of young conservation scientists in their survey. The ranking of statements by scientists in my survey reflects elements of Sandbrook et al.'s biocentric respondent grouping. Those biocentric respondents are characterized by agreement that conservation requires changes in human population growth (ranked 4 in my analyses) and must be context dependent (ranked 1 here). Respondents in Sandbrook's biocentric group disagreed that trade in wild animals could be an effective conservation tool (ranked 28 here) and that the value of biological diversity depends on its usefulness to people (ranked 31 here).

Statement rankings (Table 2) in my study identified some potentially controversial topics. For example, the statement ranked ninth overall in times selected as most agreed with (“Science should be used to determine—not simply inform—policy.”) was also chosen as the least agreed with statement 24.0% of the times it was shown. It is not until one reaches the nineteenth ranked statement that there is another statement that garners as many negative rankings.

Management and Policy Opinions

I highlight 3 points from the overall responses to statements regarding potentially controversial conservation strategies. First, discussion of the concept of triage has long been considered off limits among some conservation scientists (Marris 2007). Results from my survey demonstrate, however, that many scientists are potentially supportive of triage and prioritization efforts. For example, 50.3% and 9.3% of scientists agree or strongly agree, respectively, with the statement “Species and ecosystems are going to unravel so it is important that the conservation community considers criteria for triage decisions. If we don’t, ad hoc decisions could be even worse.”Hagerman et al. (2010) also recently found a willingness among conservation professionals to openly discuss triage issues. They argue it is time to move beyond outright rejection of triage. Results from my survey suggest that a shift in attitude may have already happened or that it always existed.

Second, there seemed to be relatively modest agreement among respondents on the need to integrate ecology and economic analyses (41.5% agree or strongly agree, 26.2% neither agree nor disagree, 32.3% disagree or strongly disagree with the statement, “Economic valuation of species and ecosystems is essential for better societal decision-making.”) and some skepticism regarding the feasibility of integrated analyses (56.2% agree or strongly agree, 27.6% neither agree nor disagree, 16.2% disagree or strongly disagree with “Commoditization of species and ecosystems is inherently dangerous because it does not, and cannot, consider irreplaceable functions of biological diversity.” 47.4% agree or strongly agree, 18.5% neither agree nor disagree, 33.1% disagree or strongly disagree with “Biologists and economists cannot realistically link ecosystem function to economic value. This is a major weakness with current widespread adoption of the ecosystem services framework.”). Treating species and ecosystems as commodities was generally viewed negatively (56.2% agree or strongly agree, 27.6% neither agree nor disagree, 16.2% disagree or strongly disagree with “Commoditization of species and ecosystems is inherently dangerous because it does not, and cannot, consider irreplaceable functions of biological diversity.”), and respondents expressed relatively heavy support for rules and enforcement (64.1% agree or strongly agree, 21.3% neither agree nor disagree, 14.6% disagree or strongly disagree with “We need more rules, better monitoring, increased enforcement, and larger fines. Making damaging human behavior illegal and expensive is central to any strategy meant to protect biological diversity.”). In addition, there was a relatively high level of agreement that the species with highest probabilities of extinction and ecosystems with highest probability of land-cover conversion should receive the highest levels of investment (41.7% agree or strongly agree, 31.4% neither agree nor disagree, 26.9% disagree or strongly disagree with “The most vulnerable species and ecosystems should receive the highest levels of investment precisely because of their vulnerability.”). This is potentially in contrast to supporting comparison of marginal costs and benefits to guide investment decisions. Seen in conjunction with results from Sandbrook et al.'s (2011) ranking exercise, these results suggest scientists do not fully support the ecosystem-services concept (TEEB 2010).

Third, the majority of respondents agreed or strongly agreed that conservational professionals need to be willing to rethink conservation goals and standards of success (82.0% agree or strongly agree, 13.7% neither agree nor disagree, 4.3% disagree or strongly disagree with “Conserving biological diversity in an era of climate change means that conservation professionals need to be willing to rethink conservation goals and standards of success.”). Hagerman et al. (2010) noted that climate change offered an opportunity to expand conservation goals and potentially transform conservation policy. They point out substantive change can take time and may require the critical mass of a new generation. Results from my survey suggest there is, however, already widespread belief that substantive change in conservation goals is needed. Five clusters differentiated scientists’ views on controversial conservation strategies. The interventionist cluster was central to the division of opinions on the basis of scientific standing; scientists with h≥ 13 were more likely than those with h < 13 to belong to the interventionist segment and less likely to belong to the naturally oriented cluster. Senior scientists may be more open than junior scientists to redefining conservation goals. The paucity of other significant covariates suggests scientists’ core values are driving their opinions regarding preferred management options and that those values are not reflected in their demographic characteristics or professional training.

The key message of my results is that there is overwhelming agreement on the overall extent and geographic scope of loss of biological diversity among scientists with diverse professional and demographic characteristics. The degree of consensus regarding the loss of biological diversity is, in fact, much higher than the degree of consensus for the existence of anthropogenic climate change among climate scientists (Rosenberg et al. 2010). It may soon be possible to assess whether scientists’ opinions on the magnitude and timing of loss coincide with new scenario models of loss of biological diversity (Pereira et al. 2010).

The degree of consensus on the magnitude of current and projected losses of biological diversity may increase policy makers’ level of confidence that investments in scientific modeling, monitoring, restoration, and institutional reform are warranted. Given the perceived severity of loss of biological diversity, scientists may be willing to discuss potentially contentious conservation options. A willingness to engage in wide-ranging discussions of these options could give policy makers more ideas and latitude with regard to conservation issues. It seems particularly timely that now, as Conservation Biology celebrates its 25th anniversary, we could be on the cusp of a period of evolution in thinking about how conservation goals might be redefined and realized as the effects of human activities and climate change escalate rapidly.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

I sincerely thank survey respondents, who took time from their busy schedules to complete the rather complicated survey. I also thank three reviewers and editorial staff for valuable comments on the draft manuscript.

Literature Cited

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information
  • Anderegg, W. R. L., J. W. Prall, J. Harold, and S. H. Schneider. 2010. Expert credibility in climate change. Proceedings of the National Academy of Sciences of the United States of America 107:1210712109.
  • Dillman, D. A., J. D. Smyth, and L. M. Christian. 2009. Internet, mail, and mixed-mode surveys: the tailored design method, 3rd edition. Wiley, Hoboken , New Jersey .
  • Eid, M., R. Langeheine, and E. Diener. 2003. Comparing typological structures across cultures by multigroup latent class analysis: a primer. Journal of Cross-Cultural Psychology 34:195210.
  • Flynn, T. N., J. J. Louviere, T. J. Peters, and J. Coast. 2007. Best-worst scaling: what it can do for health care research and how to do it. Journal of Health Economics 26:171189.
  • Hagerman, S., H. Dowlatabadi, T. Satterfield, and T. McDaniels. 2010. Expert views on biodiversity conservation in an era of climate change. Global Environmental Change 20:192207.
  • Harzing, A.-W. 2010. The Publish or Perish Book. Tarma Software Research, Melbourne , Australia .
  • Hirsch, J. E. 2005. An index to quantify an individual's scientific research output. Proceedings of the National Academy of Sciences of the United States of America 102:1656916572.
  • Jones, J. P. G., et al. 2011. The why, what, and how of global biodiversity indicators beyond the 2010 target. Conservation Biology 25: 450457.
  • Lubchenco, J. 1998. Entering the century of the environment: a new social contract for science. Science 279:491497.
  • Magidson, J. 2005. SI-CHAID 4.0 User's guide. Statistical Innovations, Belmont , Massachusetts .
  • Magidson, J., and J. K. Vermunt. 2005. An extension of the CHAID tree-based segmentation algorithm to multiple dependent variables. Pages 176183 in C. Weihs and W. Gaul, editors. Classification: the ubiquitous challenge. Springer, Heidelberg , Germany .
  • Marris, E. 2007. What to let go. Nature 450:152155.
  • Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being: synthesis. Island Press, Washington , D.C .
  • Morgan, M. G., H. Dowlatabadi, M. Henrion, D. Keith, R. Lempert, S. McBride, M. Small, and T. Wilbanks, editors. 2009. Best practice approaches for characterizing, communicating, and incorporating scientific uncertainty in climate decision making. National Oceanic and Atmospheric Administration, Washington , D.C .
  • Olson, D. M., and E. Dinerstein. 1998. The Global 200: a representation approach to conserving the Earth's most biologically valuable ecoregions. Conservation Biology 12:502515.
  • Pereira, H. M., et al. 2010. Scenarios for global biodiversity in the 21st century. Science 330:14961501.
  • Perrings, C., A Duraiappah, A. Larigauderie, and H. Mooney. 2011. The biodiversity and ecosystem services science-policy interface. Science 331:11391140.
  • Rosenberg, S., A. Vedlitz, D. Cowman, and S. Zahran. 2010. Climate change: a profile of US climate scientists’ perspectives. Climatic Change 101:311329.
  • Rudd, M. A. 2011. How research-prioritization exercises affect conservation policy. Conservation Biology 25:860866.
  • Sandbrook, C., I. R. Scales, B. Vira, and W. M. Adams. 2011. Value plurality among conservation professionals. Conservation Biology 25:285294.
  • Sawtooth Software. 2007. The MaxDiff/Web 6.0 technical paper. Sawtooth Software, Sequim , Washington .
  • SCBD (Secretariat of the Convention on Biological Diversity). 2010. Global biodiversity outlook 3. SCBD, Montréal .
  • TEEB (The Economics of Ecosystems and Biodiversity). 2010. Mainstreaming the economics of nature: a synthesis of the approach, conclusions and recommendations of TEEB. TEEB, United Nations Environment Programme, Bonn .
  • Vermunt, J. K., and J. Magidson. 2002. Latent class cluster analysis. Pages 89106 in J. A. Hagenaars and A. L. McCutcheon, editors. Applied latent class analysis. Cambridge University Press, Cambridge , United Kingdom .
  • Vermunt, J. K., and J. Magidson. 2005. Technical guide for Latent GOLD 4.0: basic and advanced. Statistical Innovations, Belmont , Massachusetts .

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

Appendix S1

Appendix S2

Appendix S3

Appendix S4

Appendix S5

Appendix S6

FilenameFormatSizeDescription
COBI_1772_sm_SuppMat1.doc324KSupporting info item
COBI_1772_sm_SuppMat2.pdf354KSupporting info item
COBI_1772_sm_SuppMat3.pdf300KSupporting info item
COBI_1772_sm_SuppMat4.pdf87KSupporting info item
COBI_1772_sm_SuppMat5.xls1110KSupporting info item
COBI_1772_sm_SuppMat6.xls347KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.