Reply to Comment on “Scientists’ Views about Attribution of Global Warming”

e thank JoseDuarte for his interest in our paper. Thecriticisms in his comment are 3-fold: (1) he claimed weincluded an unknowable number of “non-climate scientists” inour survey; (2) the inclusion of impacts and mitigationresearchers biases our results on the level of consensus upward;and (3) there is pressure to abide by the consensus, precludingany conclusions to be drawn from its existence. In response, weargue that the number of “non-climate scientists” in our surveyis known to be small and their in- or exclusion does not changeour conclusion that the level of consensus increases withincreasing expertise. With respect to point 2, we reiterate thatwe intended to survey the wider scientific field that works onclimate change issues. This has actually led to a slightly lowerlevel of consensus than if we had only surveyed physical climatescientists. Finally, Duarte’s characterization as if a scientificconsensus is somehow enforced by nefarious means lackssubstantiation.Survey respondents were asked for the number of years thata respondent had been professionally involved with climatechange issues (Q7a) and for the number of climate-relatedarticles written in the peer-reviewed literature (Q7b).Researchers from a nonclimate related field, who were admittedto our survey because they wrote an article with the keyword“global climate change” or “global warming”, would haveanswered zero to one or both questions, assuming theyanswered truthfully. The size of this group of “non-climatescientists” in our survey is 81 (∼4% of the respondents). If theywere excluded from our survey, the level of concensus based onQ1 of our total group of respondents who expressed anopinionthat is, excluding the undetermined responseswould remain the same: 84%.Duarte argues that inclusion of scientists with self-reportedexpertise in climate impacts (WG2) or mitigation (WG3) causean inflationary bias for the level of consensus found. However,we consciously included these professionals in our survey setup.In our article we describe how respondents were selected andthat we surveyed “scientists studying various aspects of climatechange, including physical climate, climate impacts, andmitigation.” This setup of surveying the wider scientific fieldof climate-related researchers enabled us to investigate in detailhow the various views on climate science depend on differentmetrics related to expertise.Table S3 (SI) provides the consensus results for severalsubgroups, including those with self-declared WG1 expertise(physical climate scientists), WG2 and WG3. Those with WG1expertise indeed reported a slightly lower consensus level thanthose with WG2 or WG3 expertise. However, the level ofconsensus of those with WG1 expertise is slightly higher thanthat of the total group of respondents, since the latter alsoincludes respondents who did not report any of the IPCCWG1, WG2, or WG3 expertise fields, of whom 78% agrees withthe IPCC attribution statement. Moreover, the 88 signatories ofpublic statements critical of climate sciencethe“uncon-vinced”exhibit a very low consensus level of 12%. So theselatter two groups “bias” the consensus level in the oppositedirection and probably contributed to us finding a lower level ofconsensus than several other studies did.Our main conclusion, that the level of consensus increaseswith increasing expertise in climate science, as defined by thenumber of self-reported articles in the peer-reviewed literatureon climate change, is not affected by having included impactsand mitigation researchers, or other tangentially relatedprofessionals.Duarte bases his claim that dissent in climate science isoppressed on a few anecdotes and innuendo. He uses wordssuch as “McCarthyite ”, smearing , and savaged , but onlymentions anecdotes that support his case, seemingly unaware ofthe many anecdotes in which self-proclaimed skeptics engage insuch behavior toward mainstream scientists. A scientist who isskeptical of the anthropogenic causation of climate change maybe called a “denier”; a scientist who is convinced that climatechange is predominantly human-induced may be called an“alarmist”. Judged by the examples he draws upon, his opinionappears strongly influenced by the skeptical blogosphere, whichis rife with unsupported accusations of wrongdoing.These sorts of anecdotes exemplify the polarized nature ofthe public debate on climate change, which is to an extent dueto the “wicked” nature of the problem and the moral aspectsthat it touches upon. Duarte’s claim that dissent is oppressedhas a conspiratorial tone to it. As such it cannot be disproven,but it can be pointed out that it lacks real-world evidence.Duarte describes how psychological driversfear ofostracismcould lead scientists to pay lip-service to theconsensus, even if they do not fully agree. However, making aType I error (false positive or false alarm) is probably moredamaging to a scientist’s reputation and credibility than makinga Type II error (false negative or missed opportunity).