EGU2020-7302
https://doi.org/10.5194/egusphere-egu2020-7302
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Estimating global mean sea-level rise and its uncertainties by 2100 and 2300 from expert assessment

Benjamin Horton1, Nicole Khan2, Niamh Cahill3, Janice Lee1, Tim Shaw1, Andra Garner4, Andrew Kemp5, Simon Engelhart6, and Srefan Rahmstorf7
Benjamin Horton et al.
  • 1Nanyang Technological University, Asian School of the environment, Singapore, Singapore (bphorton@ntu.edu.sg)
  • 2Department of Earth Sciences and Swire Marine Institute, The University of Hong Kong, Hong Kong.
  • 3Department of Mathematics and Statistics, Maynooth University, Maynooth, Ireland.
  • 4Department of Environmental Science, Rowan University, Glassboro, NJ 08028
  • 5Department of Earth and Ocean Sciences, Tufts University, Medford, MA 02155, USA
  • 6Department of Geography, Durham University, Durham, DH1 3LE, UK
  • 7Potsdam Institute for Climate Impact Research, Telegrafenberg A62, 14473 Potsdam, Germany.

Sea-level rise projections and knowledge of their uncertainties are vital to make informed mitigation and adaptation decisions. To elicit expert judgments from members of the scientific community regarding future global mean sea-level (GMSL) rise and its uncertainties, we repeated a survey originally conducted five years ago. Under Representative Concentration Pathway (RCP) 2.6, 106 experts projected a likely (at least 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 (i.e. an exceedance probability of 17%) range estimated by the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. 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.

How to cite: Horton, B., Khan, N., Cahill, N., Lee, J., Shaw, T., Garner, A., Kemp, A., Engelhart, S., and Rahmstorf, S.: Estimating global mean sea-level rise and its uncertainties by 2100 and 2300 from expert assessment, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-7302, https://doi.org/10.5194/egusphere-egu2020-7302, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 03 May 2020
  • CC1: expert overconfidence, medians, and probabilities, Aslak Grinsted, 08 May 2020


    Thank you for doing this work. One thing I have been wondering about since your previous survey is how "the overconfidence bias" would affect the estimated ranges. What I mean when i say "the overconfidence bias": 

    * Capen 1976 found an almost universal tendency to underestimate uncertainty. 

    * “Oil & Gas exploration personnel are commonly required to give 80% ranges to describe parameters of a potential oil field and, historically, these ranges have been too narrow, precisely as the overconfidence bias predicts. Hawkins et al (2002), for example, report that ‘actuals’ fall outside the predicted 80% ranges more than half the time” -Welsh

    I believe you used median values for the final ranges in your paper. But if we know that most expert are overconfident, and have too narrow uncertainties, then the median expert would also tend to be overconfident. In my teaching I have tried to replicate the old capen quiz (in a non-scientific manner), and on these data the median range seem to be too narrow. For this reason it seems to me that it is difficult to assign probabilities to the ranges derived in this manner.

    Do you have any thoughts on this? 

    • AC1: Reply to CC1, Benjamin Horton, 09 May 2020

      Good point and it is important that you raised the issue of the overconfidence bias.
       We are aware of the potential problem of overconfidence bias. Indeed this is one of the differences between a broad elicitation that is used in our paper and structured expert judgments, which were used in the recent survey paper by Jonathan Bamber regarding ice sheet contribution to SLR.
      Broadly, two different approaches have been used to extract estimates of the likelihood of future sea-level rise from experts in the field. Structured expert judgment is a formal method in which a small number of experts are guided in the interpretation of probabilities in a workshop setting before having their responses weighted based on their performance on calibration questions. More informal approaches, also known as “broad” elicitations or expert (surveys), ask many experts a small number of questions, aiming for wide participation by minimizing the required time investment for participation. 

      We previously used broad elicitation in our paper in 2014, so to allow comparison we kept our survey the same. We made sure in the paper that we were very clear what type of survey this was.