Read on Twitter
Two important studies out in @nature yesterday projecting future sea level rise from Antarctica and other sources which are likely to drive discussion among ice sheet modelers, so a 
with some thoughts and aggregated resources (1/n)
with some thoughts and aggregated resources (1/n)
The first (led by @robdeconto) updates Antarctic Ice Sheet projections from a previous study, with new constraints, new emulation, new emissions scenarios and model improvements (2/n) https://doi.org/10.1038/s41586-021-03427-0
The other (led by @flimsin) has new probabilistic projections of sea level rise to 2100, integrating results from recent model intercomparison projects using a series of Gaussian Process emulators (3/n) https://doi.org/10.1038/s41586-021-03302-y
There are a number of great threads from co-authors and news stories on the headline findings, which I'll link at the bottom of this thread. Here, Iâll point out some of the bits I found the most interesting and how they point towards future research priorities (4/n)
@robdeconto et al. discuss work by B. Parizek and F. Clerc (with @MITglaciology) finding a wide range of maximum ice cliff heights supportable under shear failure depending on the assumed shear strength of damaged ice (5/n) https://doi.org/10.1029/2019GL084183 https://doi.org/10.1130/G45880.1
There is a lack of consensus on cliff heights because the shear strength of damaged glacier ice is difficult to measure, but there are promising new approaches (e.g. from @LizzUltee ) using observations to infer these important parameters (6/n) https://doi.org/10.1017/jog.2020.65
The best constraint on the calving rate at tall ice cliffs comes from Sermeq Kujalleq (aka Jakobshavn Isbrae) which retreated very rapidly from 2000 to 2015, though recently it has re-formed a small ice shelf with local ocean cooling (7/n) https://doi.org/10.1038/s41561-019-0329-3
Recent work has started to disentangle the effects of ocean melt, melange and calving at the terminus of SK, though there is still no consensus on the main drivers of retreat (8/n) https://doi.org/10.1029/2018GL079827 https://doi.org/10.5194/tc-14-211-2020
The only way to produce tall ice cliffs (i.e. which don't think viscous deformation) is the rapid collapse of ice shelves, which in this model occurs with enough surface melt driving hydrofracture (9/n)
Recent work from @JKingslake @chingyaolai and @Dr_Gilbz look at how much surface melt occurs on ice shelves (now and in the future) and whether water is added or removed from regions prone to fracture (10/n) https://doi.org/10.1038/nature22049 https://doi.org/10.1038/s41586-020-2627-8 https://doi.org/10.1029/2020GL091733
@AliBanwell and I showed that it is difficult to produce hydrofracture-driven ice shelf collapse on very short time scales (i.e. weeks), except... (11/n) https://doi.org/10.1029/2019GL084397
...when intense melt events occur, likely due to an atmospheric river, which seem to be increasingly important drivers of Antarctic surface melt as shown by @JonathanWille (though not well captured by climate models) (12/n) https://doi.org/10.1038/s41561-019-0460-1
The addition of modern and improved paleo-constraints and other model improvements leads to a median projection for 2100 SLR that is less than half the previous projection from this team (13/n) https://doi.org/10.1038/nature17145
Rapid calving at tall ice cliffs is necessary to fit paleo-constraints, the opposite conclusion of the 2019 study by @flimsin and colleagues, largely due to the addition of new higher Pliocene sea level constraints from recent studies (14/n) https://doi.org/10.1038/s41586-019-1619-z
This back and forth on what the paleo-constraints actually tell us about "necessary" ice sheet processes underlines the importance of continuing to narrow paleo-sea level estimates, as was shown nicely by @danielgilford in a recent paper (15/n) https://doi.org/10.1029/2019JF005418
None of the constraints in this study (modern or paleo) rule out any simulations on the upper bound of the ensemble, indicating that much higher potential sea level rise from Antarctica is possible, though not simulated here (16/n)
...which raises the question of how we should design the prior (finite support?) in ensemble studies in the absence of high-end constraints, though @glacierandy et al. have done excellent work in this direction for Greenland projections (17/n) https://doi.org/10.1126/sciadv.aav9396
The new study from @flimsin et al provides an interesting contrast largely because it provides probabilistic projections without constraints from sea level (though of course the models within it are validated on modern glaciological observations) (18/n)
The ISMIP6 ensemble emulated in the @flimsin study does not rule out the zero or small positive SLR contribution from Antarctica in 2100, though @robdeconto et al show this can be ruled out solely on the basis of modern sea level constraints (19/n) https://doi.org/10.5194/tc-14-3033-2020
2100 is exactly when many Antarctic models in ISMIP6 are beginning rapid ice loss, while others have changed little, leading to drastically divergent projections, which weâve shown is an expected property of unstable systems (20/n) https://doi.org/10.1073/pnas.1904822116
âRisk-averseâ projections seem to be particularly useful for stakeholders who do not necessarily want to parse the differences between models and emissions scenarios, where risk-averse sea level decisions are very sensitive to high-end uncertainty (21/n) https://doi.org/10.1007/s10584-016-1664-7
What I find so exciting about both studies is that they touch on so many of the interesting outstanding questions within our field and provide concrete connections from these fascinating glaciological puzzles to real consequences for coastal communities (22/n)
Anyway, just some of the great news stories on these studies which get the nuance right https://www.nytimes.com/2021/05/05/climate/climate-change-sea-level-rise.html
https://www.washingtonpost.com/climate-environment/2021/05/05/uncertainty-is-not-our-friend-scientists-are-still-struggling-understand-sea-level-risks-posed-by-antarctica/ https://www.theguardian.com/environment/2021/may/05/antarctica-ice-sheet-melting-global-heating-sea-level-rise-study
https://www.washingtonpost.com/climate-environment/2021/05/05/uncertainty-is-not-our-friend-scientists-are-still-struggling-understand-sea-level-risks-posed-by-antarctica/ https://www.theguardian.com/environment/2021/may/05/antarctica-ice-sheet-melting-global-heating-sea-level-rise-study
And a great description of the research process from @flimsin https://www.theguardian.com/commentisfree/2021/may/06/stop-rising-sea-levels-scientists-climate-forecast
Explainers/threads from @CarbonBrief https://twitter.com/CarbonBrief/status/1389960136496750593
