I believe that the linked reference contains a reasonable discussion, for the public, of many of the risks of abrupt sea level rise this century (see the attached extracts and images with associated captions). Nevertheless, as it is doubtful that any state-of-the-art Earth System climate models (like ACME) with state-of-the-art marine glacier sub-models and associated freshwater hosing, will be adequate for characterizing the risks of abrupt sea level rise this century, for at least 10 to 20 years; I provide the following comments:
1. In the attached Figure 10 for the WAIS, the right panel roughly represents an ice cliff face retreat rate of about 1 km/year; while the left panel roughly represents an ice cliff retreat roughly 5km/year; however, Greenland marine glaciers have exhibited ice cliff retreat rates of over 10 km/year. Thus while it is laudable that DeConto et. al. have a paper in the publishing process that will examine the WAIS with ice cliff retreat rates of roughly 10 km/year; we do not know that 10 km/year is the upper limit for such a retreat rate for key WAIS marine glaciers like the Thwaites Glacier. In this regards, I note that DeConto & Pollard 2016 report an upper bound for sea level rise by 2100 of 2.5 +/- 0.15m; Hansen indicates that there is a risk that global mean sea level rise might reach 5 m SLR by 2100.
2. While model makers need to do the best that they can; policy makers should conduct probable failure mode analyses, PFMA, to try to attempt to characterize the risks not captured by current climate models. In this regards, I note that the paleo record for the MIS 11 (Holsteinian) era indicates a climate sensitivity much higher than current ESMs can match, and this many well be because they do not adequately model Hansen's ice-climate feedback mechanism.
3. Examples of mechanism's not yet adequately characterized by current state of the art WAIS ice sheet models (besides the 10km/year ice cliff retreat rate, yet to be published by DeConto et. al) include: (a) possible ice cliff retreat rates higher than 10km/year; (b) the influence of the high geothermal heat flux beneath all WAIS marine glaciers; (c) potential abrupt ice mass loss triggered local earthquakes, volcanic eruptions and methane emissions from seafloor hydrates; d) bipolar interactions between the GIS and the AIS; and (d) the influence of increased local storm activity associated with Hansen's ice-climate interaction. Also, I note that the Antarctic folder identifies numerous other possible positive feedback mechanisms not yet addressed by current ESM projections (including high values of ECS):
Griggs, G, Arvai, J, Cayan, D, DeConto, R, Fox, J, Fricker, HA, Kopp, RE, Tebaldi, C, Whiteman, EA (California Ocean Protection Council Science Advisory Team Working Group). Rising Seas in California: An Update on Sea-Level Rise Science. California Ocean Science Trust, April 2017.
http://www.opc.ca.gov/webmaster/ftp/pdf/docs/rising-seas-in-california-an-update-on-sea-level-rise-science.pdfExtract: "In the future ice-sheet ensembles shown in Figure 10, a range of maximum cliff-failure rates are used, ranging between one and five km per year. At the tallest vertical ice cliffs observed today (e.g., Helheim and Jakobsavn glaciers in Greenland), the horizontal rate of cliff retreat is as high as 10-14km per year (Joughin et al., 2010; 2012). This is quite remarkable, considering these outlet glaciers rest in narrow fjords 5 to 12 km wide, choked with dense melange as seen in Figure 9.
In Antarctica, the cliff faces that could appear in the future will be much taller and wider than those in Greenland, where melange can clog seaways. For example, Thwaites Glacier is >120 km wide and its terminus ends in open ocean rather than a narrow fjord, so it might be reasonable to assume cliff collapse in open settings like Thwaites could approach the rates observed in narrow Greenland fjord settings where melange is presumably providing some back pressure at the grounding line. Increasing the model’s maximum cliff retreat values closer to those observed in Greenland (~10 km per year) increases Antarctica’s simulated contribution to GMSL to more than 2m by 2100 in the RCP8.5 scenario (DeConto et al., in preparation)."
Captions: "Figure 8. A similar ice sheet margin as shown in Figure 6, but feeling the effects of both sub iceshelf oceanic warming and atmospheric warming. Meltwater and rainwater accumulating on the iceshelf surface can fill crevasses (a), which deepens the crevasses, potentially leading to hydrofracturing (b). If the newly exposed grounding line is thick enough to have a tall subaerial ice cliff (c), the terminus would fail structurally. If the rate of structural failure outpaces the seaward flow of ice, the ice margin would back into the deep basin (after Pollard et al., 2015; DeConto and Pollard, 2016), resulting in amassive loss of ice."
"Figure 10. Ensembles of Antarctic’s future contribution to sea level, using paleo-calibrated ice-model physics, high resolution atmospheric climatologies from a regional atmospheric model, and time-evolving ocean model temperatures (from DeConto and Pollard, 2016). The inset at right shows time-evolving CO2 concentrations (RCPs) used to force the ice sheet simulations (from van Vuuren et al., 2011). Note that different colors are used to represent the RCPs and ice sheet ensembles. The difference between the ensembles at left versus right lies in the assumptions used in the model calibration (based on geological sea-level reconstructions). These differences demonstrate the large uncertainty remaining in current projections. The timing when Antarctica begins major retreat in RCP4.5 and 8.5 (after ~2060) also remains uncertain. In addition to greenhouse-gas forcing, the onset of major retreat will be dependent on the trajectory of Antarctic warming in response to a complex combination of factors including recovery of the ozone hole, linkages with tropical dynamics, and feedbacks between the ice-sheet, solid-Earth, ocean, and sea ice which are not accounted for here. Addressing these shortcomings and uncertainties will be the focus of future work."
"Figure 12. Porous firn (a, left) can absorb seasonal meltwater and delay water flow into underlying crevasses (b, right), delaying hydrofacturing and ice-shelf breakup. Better treatments of these processes in ice sheet models will be critical for predicting the precise timing of the ice sheet’s response to a warming climate (figure source: Munneke, et al., 2014)."