Public overestimates sea-level rise possible by 2100

We first examined New Zealanders’ understanding of current sea-level rise. As noted, average global sea level has risen by more than 16 cm since 1900 [1]. Respondents were thus asked to identify, from options ranging from ‘not at all’ to ‘more than 40 cm’, how much they thought global sea level has risen since 1900. More than a quarter of respondents answered ‘10–20 cm’ (26.9%), a figure that is in line with current understanding of sea-level rise to date. But a larger proportion (38.2%) of respondents either underestimated the amount of sea-level rise, selecting ‘not at all’ (11.2%) or ‘5–10 cm’ (27%), or overestimated (35%) the amount of sea-level rise from 1900 to date, selecting ‘20–30 cm’ (13.4%), ‘30–40 cm’ (8.4%) or ‘more than 40 cm’ (13.2%).

We then examined knowledge of sea-level rise projections to 2100 (see Fig 1A). While a minority (6.8%) of respondents said that sea level was not projected to rise by 2100, nearly 75% of respondents selected options that were in line with scientifically-plausible projections, selecting ‘up to 40 cm’ (28.6%), ‘up to 1 m’ (30.9%), and ‘up to 2 m’ (14.9%). But a large group of respondents (18.9%) overestimated sea-level rise projections to 2100, selecting ‘up to 5 m’ (10.7%) or ‘more than 5 m’ (8.2%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Sea-level rise projections to 2100.

(a): Respondents were asked ‘How much do you think global sea level is projected to rise by 2100?’ (b): Respondents were asked ‘How much could global average sea level rise under ‘a scientifically credible worst-case scenario’?.

https://doi.org/10.1371/journal.pone.0254348.g001

To further explore the amount of sea-level rise respondents thought was possible by 2100, we asked how much global sea level could rise by 2100 under ‘a scientifically credible worst-case scenario’ (see Fig 1B). While a similar small number of respondents said that sea-level rise was not going to happen at all (7%), and 22.5% of respondents underestimated the worst-case scenario by selecting ‘up to 1 metre’, this time only 33.1% of respondents gave an answer broadly in line with current science, answering ‘1 m or more’ (16.7%) or ‘2 m or more’ (16.4%). Another 37.4% of respondents substantially overestimated sea-level rise by 2100, selecting ‘5 m or more’ (18%), ‘8 m or more’ (7.3%), ‘12 m or more’ (5.3%) or ‘15 m or more’ (6.8%).

To assess what respondents thought was physically possible, we then asked what they thought would be the ‘maximum amount of sea-level rise’ that could occur if ‘all of the glaciers and ice sheets and ice caps on planet Earth melted’, and what was the fastest time period over which this could occur. The largest group (38.8%) of respondents underestimated the maximum amount of sea-level rise possible, selecting ‘about 30 m’. The next largest group (22.3%) chose a figure broadly in line with current estimates of 66.07 m [29], selecting ‘about 60 m’. The remaining 38.9% of respondents overestimated total maximum sea-level rise, selecting ‘about 120 m’ (17%), ‘about 240 m’ (10.9%) and ‘more than 500 m’ (11%).

In selecting the ‘fastest period of time’ over which their perceived maximum amount of sea-level rise could occur, respondents overestimated the speed at which all the planet’s ice could melt. While scientists suggest that such melting could take place under a sustained warming climate over a period of thousands of years [30, 31], the majority of respondents overestimated how fast the planet’s ice could melt, selecting ‘decades’ (33.1%) or ‘centuries’ (39%). A smaller group (11.6%) of respondents answered in line with the science by selecting ‘thousands of years’. The remaining respondents selected ‘tens of thousands’ (3.2%) or ‘hundreds of thousands’ (2.7%) of years, or acknowledged that ‘I really don’t know’ (10.4%).

Public identified melting sea ice as causing sea-level rise

Respondents were also asked to identify and rank the major causes of sea-level rise from a ten-item list that included melting ice sheets, melting land-based glaciers, melting sea ice, thermal expansion, and land subsidence. No definitions of these phrases were provided as our intention was to gauge people’s understanding of sea-level rise from existing sources of information, such as the media. While climate scientists and the media make efforts when communicating with publics to not identify melting of Arctic and Antarctic sea ice (which is widely covered in the media) as causing sea-level rise, 28.7% of respondents identified this as their top ranked cause of sea-level rise (see Fig 2A). Encouragingly, other respondents’ top ranked causal mechanisms that do contribute to sea-level rise, selecting melting ice sheets (24%), thermal expansion of the oceans (7%), and melting of land-based glaciers (4.2%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Major contributors to sea-level rise.

Respondents were asked to identify and rank the major contributors to sea-level rise. In (a) we show respondents’ top ranked cause of sea-level rise (causes selected by fewer than 2% of respondents are not shown). In (b) we show the sum of the top three causes identified by respondents (causes ranked by fewer than 10% of respondents are not shown).

https://doi.org/10.1371/journal.pone.0254348.g002

When we grouped respondents’ top three rankings for causes of sea-level rise (Fig 2B), melting ice sheets (62.3%) was first, followed by melting sea ice (59.8%), melting of land-based glaciers (44.1%), thermal expansion of the oceans (23.6%) and land subsidence in some areas (14.2%). There were also large proportions of respondents ranking some other mechanisms that do not impact on sea-level rise, such as ‘increased slash and sediment runoff due to deforestation’ (14.3% ranked it in their top three), ‘dumping of rubbish/waste into the ocean’ (14%), ‘increased rainfall in the tropics’ (9.9%) and ‘submarine landslides’ (7.5%).

Notably, 25.9% of respondents also ranked ‘something else’ as one of the top three causes of sea-level rise. When asked to specify what these causes might be, more than half (54%) were unsure or gave no answer, while 23.5% attributed sea-level rise broadly to climate change, global warming or human activities, and 16.5% gave answers such as ‘solar activity’, ‘earthquakes’, or ‘storms’, suggesting these respondents believe natural processes are responsible for sea-level rise.

Challenges of projecting and communicating sea-level rise

Projecting sea-level rise is complicated [32]. Different climate, mountain glacier and polar ice sheet models produce different results. This is especially the case for Antarctic ice sheet models, which can range from projecting a net gain in mass to a significant loss. Thus, the IPCC approach is to take the mean of ensembles of model experiments, that all use broadly similar boundary conditions and climate forcings. Moreover, glacier melting, vertical land movement, gravitational effects, and ocean dynamics all contribute to sea-level rise being different in different locations; and calculations must be made for a number of different scenarios representing different mitigation pathways [33]. Sea-level rise projections are therefore usually scenario dependent and framed in terms of a probablistic range, rather than a single figure, to reflect the uncertainty in the projections.

Communicating about sea-level rise, given the uncertainties inherent in projections, is also complicated [8]. IPCC reports, which assess and interpret all the latest science, contain jargon and caveats about ‘uncertainty ranges’ with evidence that publics misunderstand verbal probability expressions, known as calibrated uncertainty language [34], such as ‘likely’ and ‘unlikely’ [35]. Even the IPCC summaries for policymakers have been criticised for their formal structure and use of jargon [36]. This is a complex challenge for science communicators. As noted by Seethhaler et al, communicating about uncertainty is “one of the greatest challenges in science communication, one that raises ethical issues including, but not constrained to, finding a middle ground in prognostication between false assurances and doomsday scenarios” [37, p. 381].

Media interpretations of the science adds to the problem. While the New Zealand media gives extensive coverage to IPCC projections, it also gives voice to more extreme and catastrophic views from local and international scholars. For example, in 2016 New Zealand media reported on research that projected several metres of sea-level rise by 2100 [38–40], and in 2017 there was significant media coverage when a philosophy professor warned that sea-level rise could reach 5 m or more by 2100 [41–43]. These stories, though, also included comments from scientists who communicated the IPCC projections and from politicians who underestimated the importance of sea-level rise as an issue. Such indiscriminate media coverage, and attempts at journalistic balance, draw the uncertain nature of sea-level rise projections to the public’s attention and detract from the certainties that exist out to 2050. This could therefore be contributing to public confusion on the issue. While our pilot study revealed that the majority of respondents (68%; n = 518) felt somewhat or pretty well informed about sea-level rise, only 38% (n = 519) felt somewhat or extremely clear about the rate/timing/extent of sea-level rise to 2100.

Some of the most extreme projections selected by our respondents go beyond the 5 m of sea-level rise by 2100 that some media have reported. For example, the 6.8% of respondents who thought sea-level rise could reach ‘15 m or more’ by 2100 and the 33.1% who believed that all the planet’s ice could melt over a period of ‘decades,’ selected options that are unprecedented in the geological record and defy physical laws around how fast ice can melt, even under extreme temperature forcing [44, 45].

Beyond the more descriptive analysis of the answers, we also examined the extent to which overestimation was associated with level of concern about sea-level rise. Results from t-tests confirmed that respondents who were more likely to overestimate the amount of sea-level rise and how fast this would occur were significantly more likely to express greater concern about sea-level rise to 2100 in New Zealand, and globally, compared to their counterparts (i.e., t > 5, p < .001; Cohen’s d > .50; see Section 1.4 of the S1 File for detail).

When assessing the severity of environmental problems for geographically distant places, individuals tend to rate the severity of these problems as worse ‘there’ than ‘here’ [46–48]. Hence, we also examined whether such spatial bias would be observed in relation to concern for sea-level rise. The spatial bias effect was operationalized as the difference between ratings of national concern and global concern (i.e., average Aotearoa New Zealand minus average global). A difference score equal to zero represents judgement of no nation–global distinction, while values other than zero represent biased comparative judgements. Values above zero represent the expected spatial bias effect with global sea-level rise judged as more concerning than national sea-level rise, and values below zero represent an opposite effect. For most respondents (78.8%) a biased judgement was absent (i.e., the difference score was zero) but results from a one-sample t-test indicate the mean difference score of .08 (SD = .59) was statistically different from zero, t(1034) = 4.28, p < .001, Cohen’s d = .14. A spatial bias effect was thus observed for sea-level rise: respondents judge global sea-level rise as more concerning than national sea-level rise.

The public’s association of melting sea ice with sea-level rise, as well as its selection as the top-ranked cause of sea-level rise, provides further evidence of public confusion on this issue, which may be due to the significant media coverage given to melting sea ice in the Arctic. It is possible, though, that some respondents might have been influenced by science that says the flow of ice sheets can be modulated by the presence or absence of sea ice [49–51].

Research into how publics understand sea-level rise is still an emerging field [8] and there are few international studies with which to compare our results. However, we note that our findings that respondents tend to overestimate the amount of sea-level rise possible, contrast with a 2019 Australian study showing that New South Wales coastal users and businesses tended to underestimate the amount and rate of sea-level rise along the NSW coast with 45% of respondents predicting a rise of 1–25 cms in the next 20–50 years [52]. Similarly to our results a 2016 American study found that individuals were concerned by the impacts of sea-level rise but were uncertain about its timing [53]. This survey did not ask respondents to estimate the projected amount of sea-level rise, but asked “When do you believe the effects of sea-level rise will significantly impact the county, if ever?” Finally, a 2016 study of residents of New Jersey following Hurricane Sandy showed that while participants believed sea-level rise was occurring, they did not view sea-level rise as a current threat to themselves [54]. This survey did not ask about the projected amount of sea-level rise or its timing, but asked participants whether they believed sea-level rise was occurring and whether it was affecting them. These studies and our own show that there is plenty of space to develop more in-depth understanding of how the public perceives the threat of sea-level rise.

The present research revealed the unexpected finding that one key audience for public engagement with sea-level rise is the concerned public who overestimate sea-level rise by 2100 as being catastrophic (and scientifically implausible). While there are obvious dangers associated with a public that underestimates or minimises the dangers of sea-level rise–such as not taking adaptation measures, or not supporting governments committed to mitigation and adaptation–there are also dangers associated with a public that overestimates sea-level rise. A focus on extreme (and often unsound) projections of sea-level rise can result in public anxiety and feelings of helplessness, rather than motivation to take action to mitigate and adapt [55, 56]. Nevertheless, the present research shows that publics are aware of, and concerned about, 21^st century sea-level rise, which is already impacting on coastal dwellers and infrastructure. Any public engagement efforts will therefore need to be more focused and nuanced than raising awareness of the issue, and could include, for example, provision of ongoing and credible sea-level rise information that is accessible to a wide range of publics and decision-makers. With the knowledge that overestimating can lead to failure to act, communicating with this segment of the audience could focus on providing scientifically plausible projections of sea-level rise and communicating that both adaptation and mitigation–to avoid the most extreme sea-level rise scenarios–are necessary.