Climate ChangeRecent Post

Why is future sea level rise still so uncertain?

Three new papers in the last couple of weeks have each made separate claims about whether sea level rise from the loss of ice in West Antarctica is more or less than you might have thought last month and with more or less certainty. Each of these papers make good points, but anyone looking for coherent picture to emerge from all this work will be disappointed. To understand why, you need to know why sea level rise is such a hard problem in the first place, and appreciate how far we’ve come, but also how far we need to go.

Here’s a list of factors that will influence future regional sea level (in rough order of importance):

  • ice mass loss from West Antarctica
  • ice mass loss from Greenland
  • ocean thermal expansion
  • mountain glacier melt
  • gravitational, rotational and deformational (GRD) effects
  • changes in ocean circulation
  • steric (freshwater/salinity) effects
  • groundwater extraction
  • reservoir construction and filling
  • changes in atmospheric pressure and winds

And on top of that, the risks of coastal flooding also depend on:

  • tectonic/isostatic land motion
  • local subsidence
  • local hydrology
  • storm surges
  • tides

If that wasn’t bad enough, it doesn’t even get into why some of the bigger terms here are so difficult to constrain – but more of that below.

Meanwhile, note that the factors listed above involve the whole Earth system: the oceans, the cryosphere, the atmosphere, the solid earth and lithosphere, and a full range of scales, from the city block and shoreline, to ice dynamics that change over kilometers, to GRD footprints, to the whole global ocean. While each of these elements has a devoted scientific community, sea level rise cuts across all the disciplines. And similarly, while each of these elements has a specialized modeling capability, there is no single model that encompasses all of this (not even close – as yet).

What this means is that estimates of future sea level rise are mixes of information from multiple sources, tied together in more or less sophisticated frameworks (this is the approach in the IPCC SCROCC report and the upcoming AR6) that attempt to build a full uncertainty range from all the disparate sources of information (coupled ocean-atmosphere models, hydrology models, ice sheet models, solid earth models etc.). To reiterate, there is no ‘climate model’ prediction of global sea level rise, though the climate models we often discuss here (the CMIP-class of models), do provide some of the inputs. This means that links and feedbacks between these different elements are not always coherent – e.g. the estimates of groundwater depletion (used for irrigation) or glacier melt might not impact the soils or the freshwater budget of the downstream rivers and ocean.

Two elephant seals in the Southern Oceans arguing about marine ice cliff instability.

Yes, but what about West Antarctica?

The West Antarctic Ice Sheet (WAIS) is the elephant seal in the aquarium. Ever since the 1970s it’s been suspected that it was prone to rapid collapse because the bedrock on which it sits is below sea level (and in some places, thousands of meters below sea level). More recent research constraining Eemian sea level (~125,000 yrs ago) has confirmed that WAIS collapsed at that time, adding 3 or more meters of sea level rise to the contribution from a much reduced Greenland Ice Sheet. Moreover, present day observations from gravity sensors (GRACE/GRACE-FO) show large ice mass losses from WAIS – dominated by the rapid retreats of the Pine Island Glacier and Thwaites glacier, and concomittent decreases in ice sheet elevation (from IceSat2).

Simplified schematic of atmosphere-ocean-ice interactions (Zalasiewicz et al, 2019)

There are many interesting observations and non-observations from WAIS that make this a challenging problem. First, the melting of the ice shelves and the retreat of grounding line is being driven from below as slighty warmer circumpolar deep water (CPDW) has been pushed onto the shelf. The CPDW is thought to be affected by the shift in the westerly winds around Antarctica which have increased in recent decades due to a combination of greenhouse gas forcing and the polar ozone hole (Miller et al, 2006).

Additionally, it looks like the anomalous meltwater from WAIS is causing the local ocean to freshen, stratify and cool (see Rye et al. (2020) or Sadai et al. (2020). Both of these effects make a straightforward connection between global mean warming and WAIS mass loss tricky.

But there is more. For instance, the bedrock topography under the ice sheet is still being refined. The last major revision (BedMap2) was in 2013 (Fretwell et al., 2013), but many areas remain without good data and important revisions are still being made (Morlighem et al., 2020). Also, the topography of the ocean bottom under the ice shelves is still being discovered using autonomous underwater vehicles, for instance, under the Thwaites last year. Meanwhile Bedmap3 is underway...

Furthermore, one important factor in how WAIS will affect sea level is how fast the lithosphere will respond to changes in the ice loading (part of the GRD effects mentioned above). If the mantle is very viscous, then the response is slow and it doesn’t add much to the global sea level change. But if it’s less so, then uplift is more rapid, and it can add more SLR, faster. Unfortunately, It turns out that the specific conditions under WAIS are less viscous than was thought (Pan et al., 2021).

Recent advances

Given, then, that we don’t have a suite of models with all the effects that we can analyze to give us a measure of the uncertainty, what can we do in the meantime? First, we can analyze the models we have and estimate the structural uncertainty among them – for the processes they include. This is what Edwards et al., (2021) do. Using the ISMIP6 and GlacierMIP simulation data and a statistical emulator they map out the responses of these models to the global mean temperature change and ocean-driven melting in Greenland and Antarctica. The nice thing about this is that you aren’t tied to the emission scenarios that were initially used in the MIPs, but you can’t independently calibrate the projections to paleo-climate changes, and you are stuck with the models that were used, some of which are a little out of date.

Alternately, you can take a single ice sheet model with better calibration to paleo-climate changes and drive it with climate model-derived boundary conditions as is done by DeConto et al., 2021. This doesn’t give you an estimate of full structural uncertainty (which is high), but perhaps is more internally consistent. However, the calibration that has been done on this model is (a little) controversial, and it’s worth discussing why.

Back in 2015, Pollard et al. (2015) found that their ice sheet model was overall too stable in that it wasn’t able match the large sea level changes that have been inferred for the Pliocene 3 million yrs ago (~20 meters) Eemian 125,000 yrs ago (6 to 9 meters). They added two destabilizing mechanisms, hydrofacturing of ice shelves and something called marine-ice cliff instability (MICI) and tuned the parameters to match the target. They then used this tuned version for future projections. However, the number of potential issues in the model (or any model really) is large – from uncertainties in the bedrock topography, the boundary conditions at bedrock itself, grounding line parameterizations, the resolution, the ice rheology, the lithospheric response etc. And MICI itself is quite uncertain Clerc et al., 2020 and as Edwards et al note, no model that contributed to ISMIP6 included a MICI-like mechanism. There is no guarantee that the specific destabilizing mechanisms used were the actual mechanisms at play in the warmer period. There may be other (unexplored) variations in the ice model that could have provided as good a match and that would have different sensitivity in the modern.

To their credit, DeConto et al. have extended the calibration to Pliocene sea level, the Eemian and the rate of change observed since 1992, though the Eemian constraint is the most important. And they did vary the mantle viscosity in the sea level calculations consistent with the Pan et al values. Even better, they also explored the sensitivity to a southern ocean response to Antarctic meltwater based on Sadai et al. (2020).

The question then is whether these two approaches are consistent and/or complementary.

Total land ice contribution to SLR (Edwards et al., 2021)
Antarctic contributions to sea Level scenarios from DeConto et al (2021)

So what do they show?

As one might expect, there are a lot of moving parts in these results. Many things have been varied. But there are some notable contrasts. First off, the main results for Antarctica in Edwards et al surprisingly suggest very little sensitivity to forcing scenario – basically just a continuation of the current rates of melt, which contrasts strongly with the DeConto et al result suggesting a threshold effect by 2060 between SSP1-26 (consistent with 2ºC) and SSP2-45 (or higher). Edwards et al. also look at some more ‘low probability/high impact’ runs (their ‘simulations for the risk averse’) which are more similar to the DeConto et al. results (around 20 cm from WAIS by 2100).

Remember that the biggest uncertainty is still the emission scenario, and the higher the scenario in terms of global warming, the more uncertain the ice sheet contribution is. Another key point made by DeConto et al. is that the world doesn’t stop at 2100. The consequences of even stable temperatures post-2100, has very large long term implications for sea level. For instance, even a 2ºC eventual warming is associated with around 1 meter of SLR just from WAIS by 2300.

Work to be done

These two papers illustrate the fundamental ingredients that will (eventually) get us to a more reliable estimate of SLR. The structural uncertainty explored by Edwards et al is broad, still incomplete, but essential. The calibration against past change in DeConto et al is also essential, even if the structural uncertainty they explore is narrower. A combined approach would be enlightening – using the DeConto et al model for the current ISMIP6 protocol, and extending that project to include the Eemian as an out-of-sample test might help.

Ice sheet science and the consequent sea level rise, like many cutting-edge topics, generally has a widening of uncertainty when the tools and theory start to really kick off. It is only later that this uncertainty is constrained as more observational data is brought to bear. Then, and not before, will projections start to narrow.

Until then, the most productive way to reduce uncertainties might just be to reduce emissions.


  1. R.L. Miller, G.A. Schmidt, and D.T. Shindell, “Forced annular variations in the 20th century Intergovernmental Panel on Climate Change Fourth Assessment Report models”, Journal of Geophysical Research, vol. 111, 2006.

  2. C.D. Rye, J. Marshall, M. Kelley, G. Russell, L.S. Nazarenko, Y. Kostov, G.A. Schmidt, and J. Hansen, “Antarctic Glacial Melt as a Driver of Recent Southern Ocean Climate Trends”, Geophysical Research Letters, vol. 47, 2020.

  3. S. Sadai, A. Condron, R. DeConto, and D. Pollard, “Future climate response to Antarctic Ice Sheet melt caused by anthropogenic warming”, Science Advances, vol. 6, pp. eaaz1169, 2020.

  4. P. Fretwell, H.D. Pritchard, D.G. Vaughan, J.L. Bamber, N.E. Barrand, R. Bell, C. Bianchi, R.G. Bingham, D.D. Blankenship, G. Casassa, G. Catania, D. Callens, H. Conway, A.J. Cook, H.F.J. Corr, D. Damaske, V. Damm, F. Ferraccioli, R. Forsberg, S. Fujita, Y. Gim, P. Gogineni, J.A. Griggs, R.C.A. Hindmarsh, P. Holmlund, J.W. Holt, R.W. Jacobel, A. Jenkins, W. Jokat, T. Jordan, E.C. King, J. Kohler, W. Krabill, M. Riger-Kusk, K.A. Langley, G. Leitchenkov, C. Leuschen, B.P. Luyendyk, K. Matsuoka, J. Mouginot, F.O. Nitsche, Y. Nogi, O.A. Nost, S.V. Popov, E. Rignot, D.M. Rippin, A. Rivera, J. Roberts, N. Ross, M.J. Siegert, A.M. Smith, D. Steinhage, M. Studinger, B. Sun, B.K. Tinto, B.C. Welch, D. Wilson, D.A. Young, C. Xiangbin, and A. Zirizzotti, “Bedmap2: improved ice bed, surface and thickness datasets for Antarctica”, The Cryosphere, vol. 7, pp. 375-393, 2013.

  5. M. Morlighem, E. Rignot, T. Binder, D. Blankenship, R. Drews, G. Eagles, O. Eisen, F. Ferraccioli, R. Forsberg, P. Fretwell, V. Goel, J.S. Greenbaum, H. Gudmundsson, J. Guo, V. Helm, C. Hofstede, I. Howat, A. Humbert, W. Jokat, N.B. Karlsson, W.S. Lee, K. Matsuoka, R. Millan, J. Mouginot, J. Paden, F. Pattyn, J. Roberts, S. Rosier, A. Ruppel, H. Seroussi, E.C. Smith, D. Steinhage, B. Sun, M.R.V.D. Broeke, T.D.V. Ommen, M.V. Wessem, and D.A. Young, “Deep glacial troughs and stabilizing ridges unveiled beneath the margins of the Antarctic ice sheet”, Nature Geoscience, vol. 13, pp. 132-137, 2019.

  6. L. Pan, E.M. Powell, K. Latychev, J.X. Mitrovica, J.R. Creveling, N. Gomez, M.J. Hoggard, and P.U. Clark, “Rapid postglacial rebound amplifies global sea level rise following West Antarctic Ice Sheet collapse”, Science Advances, vol. 7, pp. eabf7787, 2021.

  7. T.L. Edwards, S. Nowicki, B. Marzeion, R. Hock, H. Goelzer, H. Seroussi, N.C. Jourdain, D.A. Slater, F.E. Turner, C.J. Smith, C.M. McKenna, E. Simon, A. Abe-Ouchi, J.M. Gregory, E. Larour, W.H. Lipscomb, A.J. Payne, A. Shepherd, C. Agosta, P. Alexander, T. Albrecht, B. Anderson, X. Asay-Davis, A. Aschwanden, A. Barthel, A. Bliss, R. Calov, C. Chambers, N. Champollion, Y. Choi, R. Cullather, J. Cuzzone, C. Dumas, D. Felikson, X. Fettweis, K. Fujita, B.K. Galton-Fenzi, R. Gladstone, N.R. Golledge, R. Greve, T. Hattermann, M.J. Hoffman, A. Humbert, M. Huss, P. Huybrechts, W. Immerzeel, T. Kleiner, P. Kraaijenbrink, S. Le clec’h, V. Lee, G.R. Leguy, C.M. Little, D.P. Lowry, J. Malles, D.F. Martin, F. Maussion, M. Morlighem, J.F. O’Neill, I. Nias, F. Pattyn, T. Pelle, S.F. Price, A. Quiquet, V. Radić, R. Reese, D.R. Rounce, M. Rückamp, A. Sakai, C. Shafer, N. Schlegel, S. Shannon, R.S. Smith, F. Straneo, S. Sun, L. Tarasov, L.D. Trusel, J. Van Breedam, R. van de Wal, M. van den Broeke, R. Winkelmann, H. Zekollari, C. Zhao, T. Zhang, and T. Zwinger, “Projected land ice contributions to twenty-first-century sea level rise”, Nature, vol. 593, pp. 74-82, 2021.

  8. R.M. DeConto, D. Pollard, R.B. Alley, I. Velicogna, E. Gasson, N. Gomez, S. Sadai, A. Condron, D.M. Gilford, E.L. Ashe, R.E. Kopp, D. Li, and A. Dutton, “The Paris Climate Agreement and future sea-level rise from Antarctica”, Nature, vol. 593, pp. 83-89, 2021.

  9. D. Pollard, R.M. DeConto, and R.B. Alley, “Potential Antarctic Ice Sheet retreat driven by hydrofracturing and ice cliff failure”, Earth and Planetary Science Letters, vol. 412, pp. 112-121, 2015.

  10. F. Clerc, B.M. Minchew, and M.D. Behn, “Marine Ice Cliff Instability Mitigated by Slow Removal of Ice Shelves”, Geophysical Research Letters, vol. 46, pp. 12108-12116, 2019.

Source link

Comment here

This site uses Akismet to reduce spam. Learn how your comment data is processed.