[LINK TO THE HOMEPAGE OF THIS SITE]

RELATED POSTS [LINK]  

[LINK]  [LINK] [LINK] [LINK] [LINK] [LINK] 

THIS POST IS A CRITICAL REVIEW OF HORTON ETAL 2020 [LINK] CITED BELOW WHERE THE AUTHORS COMPARED A SURVEY OF MEMBERS OF THE SCIENTIFIC COMMUNITY (SOMOSC) IN 2020 WITH A SOMOSC IN 2015 ABOUT CLIMATE CHANGE SEA LEVEL RISE PROJECTIONS. THESE FINDINGS ARE ALSO DESCRIBED IN A SCIENCE ALERT ARTICLE OF 11 MAY 2020 [LINK] . SIMILAR FINDINGS ARE ALSO REPORTED IN A 2018 PAPER ON SEA LEVEL RISE BY DANA NUCCITELLI [LINK] 

PART-1: CITATIONS AND ABSTRACTS OF THE TWO RESEARCH PAPERS

Horton, B.P., Khan, N.S., Cahill, N. et al. Estimating global mean sea-level rise and its uncertainties by 2100 and 2300 from an expert survey. npj Climate Atmospheric Science 3, 18 (2020). https://doi.org/10.1038/s41612-020-0121-5: Abstract: Sea-level rise projections and knowledge of their uncertainties are vital to make informed mitigation and adaptation decisions. To elicit projections from members of the scientific community regarding future global mean sea-level (GMSL) rise, we repeated a survey originally conducted five years ago. Under Representative Concentration Pathway (RCP) 2.6, 106 experts projected a likely (central 66% probability) GMSL rise of 0.30–0.65 m by 2100, and 0.54–2.15 m by 2300, relative to 1986–2005. Under RCP 8.5, the same experts projected a likely GMSL rise of 0.63–1.32 m by 2100, and 1.67–5.61 m by 2300. Expert projections for 2100 are similar to those from the original survey, although the projection for 2300 has extended tails and is higher than the original survey. Experts give a likelihood of 42% (original survey) and 45% (current survey) that under the high-emissions scenario GMSL rise will exceed the upper bound (0.98 m) of the likely range estimated by the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, which is considered to have an exceedance likelihood of 17%. Responses to open-ended questions suggest that the increases in upper-end estimates and uncertainties arose from recent influential studies about the impact of marine ice cliff instability on the meltwater contribution to GMSL rise from the Antarctic Ice Sheet. 

It is noted that in the Horton etal 2020 paper cited above, though the relatively unknown Benjamin Horton of the Asian School of the Environment is listed as lead author, the full list of authors, “Benjamin P. Horton, Nicole S. Khan, Niamh Cahill, Janice S. H. Lee, Timothy A. Shaw, Andra J. Garner, Andrew C. Kemp, Simon E. Engelhart & Stefan Rahmstorf ” includes Stefan Rahmstorf, a significant and mainline figure in climate science particularly so in climate activism and particularly so in terms of his special interest in the sea level rise issue. An analysis of the findings of this paper requires the consideration that Stefan is quite possibly the significant force behind its content. [RELATED POST ON ACTIVISM]

THE DANA NUCCITELLI PAPER:

Nuccitelli, Dana, “How much and how fast will global sea level rise?, Bulletin of the Atomic Scientists, Volume 74 Issue 3, March 4, 2018. The basic physics of how global warming contributes to sea-level rise has long been understood, but new research gives us a clearer picture of what to expect. Prior to the 1990s, most sea-level rise was attributed to melting glaciers and the thermal expansion of warming ocean waters. However, ice sheets such as those covering Greenland and Antarctica have also begun to melt and play a significant role in raising ocean levels. The author reviews the results of a number of recent studies. Two of them conclude that the seas could rise by 1M or more by the year 2100, and one argued that 1.524M or more by 2100 is a possibility. Exactly what happens, and when, will be influenced by the degree to which humans reduce carbon emissions, and the uncertain dynamics of the Antarctic ice sheet. Considerable sea-level rise will occur as a consequence of the global warming humans have already caused.

PART-2: SUMMARY OF THE SCIENCE ALERT ARTICLE [LINK] AND HORTON ETAL

Latest Estimates on Sea Level Rise by 2100 Are Worse Than We Thought
MARLOWE HOOD, AFP11 MAY 2020. THIS ARTICLE IS BASED ON HORTON ETAL 2020:  Oceans are likely to rise as much as 1.3 metres by 2100 if Earth’s surface warms another 3.5 degrees Celsius. By 2300, ice sheets covering West Antarctica and Greenland will have shed trillions of tonnes in mass. Sea levels could go up by more than 5M redrawing the planet’s coastlines according to a survey of 100 experts. About 10% of the world’s population, or 770 million people, today live less than five metres above the high tide line. Even if cap warming below 2C, the sea level could go up 2M by 2300. Earth’s average surface temperature has risen just over 1 °C since the pre-industrial era. According co-author Stefan Rahmstorf, it is clear now that previous sea-level rise estimates have been too low. The new projections for both the 2100 and 2300 horizons are significantly higher than those from the IPCC because the IPCC tends to be cautious and conservative, which is why it had to correct itself upwards several times, according to Rahmstorf. Sea-level projections in the IPCC’s 2014 Report were 60 percent above those in the previous report. While less visible than climate-enhanced hurricanes or persistent drought, sea level rise may ultimately prove the most devastating of global warming impacts. The extra centimetres of ocean water make storm surges from stronger and tropical cyclones more deadly. Across the 20th century, sea level rise was caused mainly by melting glaciers and the expansion of ocean water as it warms. But over the last two decades the main driver has become the melting and disintegrating of Earth’s two ice sheets. Greenland and West Antarctica are shedding at least six times more ice today than during the 1990s. From 1992 through 2017 they lost some 6.4 trillion tonnes in mass. Over the last decade, the sea level has gone up about 4 mm per year. By the 22nd century the waterline could rise ten times faster, even under an optimistic greenhouse gas emissions scenario. The Greenland and West Antarctic ice sheets hold enough frozen water to lift oceans about 13 metres. East Antarctica, which is more stable, holds another 50 metres’ worth.

PART-3: CRITICAL COMMENTARY ON HORTON ETAL & SCIENCE ALERT:

Sea level rise projections in the survey of experts for 2100 and 2300 under emissions scenarios RCP2.6 and RCP8.5 as reported are tabulated below in three tables. From the point of view of statistics, the essential information in forecasts of variables containing an uncertainty is the mean and the variance. However, in this case neither the mean nor the variance is provided and instead, for each of the four conditions, namely, forecasts for 2100 and 2300 for RCP2.6 and RCP8.5, two different confidence intervals are provided at 66% and 90%. The mean and variance inferred from these data are not consistent as seen in Table 1 and Table 2 below. The average of these values is therefore used and these are displayed in Table 3 below.

The net information content of this survey appears in Table 3.  It shows statistical mean forecasts for 2100 of  0.495M and 1,51M for 2100 and 2300 respectively under RCP2.6 and 1.013M and 3.998M for 2100 and 2300 respectively under RCP8.5. If the 106 expert opinions are independent, then these mean forecasts are statistically significant at α=0.05.

TABLE-1: 66% CONFIDENCE INTERVAL 

TABLE 2: 90% CONFIDENCE INTERVAL

TABLE 3: AVERAGE INFERRED MEAN AND STANDARD DEVIATION

A particular feature of the statistical significance found in the t-tests above is that it was greatly aided by the large sample size in the sample of experts that participated in the survey of experts. This analysis is therefore based on the important assumption that the 106 opinions were independent. If there is any evidence that these experts were in communication about this matter prior to the survey or that the opinion of any of these experts was influenced by the opinions of fellow experts then these data have no interpretation.

The information content of this opinion survey is that the mean values in Table 3 above tells us what the experts think will happen and the uncertainty, expressed in the survey instrument as 66% and 90% confidence intervals, and in Table 3 above as standard deviation (STDEV), tells us how unsure the experts are about their forecast. The larger the confidence intervals, the larger the variance, and the larger the variance, the greater the uncertainty, and the greater the uncertainty the less they know, such that in the limit, an infinite confidence interval contains no information and a fixed point forecast with no uncertainty contains perfect information.

As explained in a related post on the statistics of confidence intervals [LINK] , variance, standard deviation, and confidence intervals are different statistical measures of the same thing and that is the uncertainty in the estimation of the mean. Therefore, in both statistics and in information theory, the greater the variance, the larger the confidence interval, and the larger the confidence interval, the less we know. However, these variables have a unique interpretation in climate science. There, the expression of uncertainty (lack of information) in terms of large confidence intervals appears to climate science as a forecast of more extreme values that foretell even greater harm from climate change. This oddity of statistics in climate science is also discussed in a post on climate activism  [LINK] . This is the the likely explanation for the climate science preference for expressing variance as confidence intervals and the interpretation of the high end of the confidence intervals as the kind of harm we must avoid by taking climate action.

PART 4: CRITICAL COMMENTARY ON NUCCITELLI  2018

Instead of a survey of experts, the Nuccitelli paper takes a survey of “a number of recent studies” of which three had an opinion on sea level rise. Two of these studies found that sea levels could rise 1.0 meter “or more” by the year 2100. No emission scenario is mentioned. The third study said that a sea level rise of 1.524 meters or more by the year 2100 was a possibility“. From these data the author concludes that “exactly what happens (in terms of sea level rise) will depend on how diligent we are in taking climate action to reduce fossil fuel emissions. There is not much useful information in this paper except the odd note that prior to the 1990s climate scientists had thought that sea level rise comes exclusively from melting glaciers and thermal expansion of seawater but that at present (2018), scientists are also worried about sea level rise from melting ice sheets. It seems odd for a climate scientist to say that climate scientists were not aware in the 1990s that ice sheets could suffer mass loss and contribute to sea level rise given the very first sentence of the paper where it says that “The basic physics of how global warming contributes to sea-level rise has long been understood”. This odd argument that sea level rise from melting ice sheets is a new chapter in climate science not previously understood is also found in the Science Alert article.

IN CONCLUSION, we find in our analysis of the sea level rise forecasts presented in the Horton 2020 paper that the interpretation of greater uncertainty expressed as larger confidence intervals and interpreted as greater evidence of extreme sea level rise is statistically flawed. Greater uncertainty does not imply a greater importance of taking climate action. The greater the uncertainty the less we know and in the case of extreme uncertainty, we don’t really know and it is not possible to demand costly climate action by claiming ignorance. A large statistical uncertainty in the data means we don’t have enough information to demand climate action but in this paper and in climate science in general, uncertainty has a perverse interpretation such that because the greater the uncertainty the larger the confidence interval, the extreme values seen in that interval are then interpreted as  a greater urgency of climate action. The less we know the scarier it gets and the greater the need for climate action.