Sussing out sea level rise

Ocean thermal expansion, that is, the increase in water volume due to temperature alone, is relatively well understood – as is the retreat of both mountain glaciers and ice caps. While most models simulate these effectively, there is little understanding of how both the Greenland and Antarctic ice sheets will respond to climate change. This is because the full extent of ice-ocean interactions is not included in climate models – it’s no wonder when there are so many factors to consider: melting of mountain glaciers, ice caps and polar ice sheets, glacial isostatic adjustment and ocean thermal expansion to name just a few!

Mahé Perrette and his colleagues have developed a novel approach for sussing out sea level rise (SLR); combining simple models with general circulation models (GCMs) to use the benefits of both in predicting future change. To get an idea of the uncertainties associated with SLR, take a look at this graph, in which red is a highly uncertain prediction (up to 70 cm variation) and dark blue is an estimate we can be quite confident of (only 5 cm variation):

Steric changes are those that affect ocean density and dynamics through variations in temperature and salinity. Accordingly, a rise in sea level due to temperature and salinity changes is known as steric sea level rise. The magnitude of the impact of steric SLR varies widely across the globe. One of the reasons behind this is the existence of local gravity fields. The loss of ice mass due to melting means there is a lower gravitational force exerted on surrounding water masses. These changes in local gravity fields cause water to migrate away from melting ice caps.

Both the magnitude of SLR and the timescales over which it occurs varies wildly between models, but it remains clear that the effects of SLR will be felt most in the low-lying coastal regions of the world: from small island nations such as Tuvalu, which at only marginally above sea level, is a country only too aware of this threat; to the densely populated, and agriculturally important, coastal plains of Bangladesh.

Each of the contributions to SLR (meltwater from the Antarctic and Greenland Ice Sheets, and melting of mountain glaciers and icecaps) vary locally, which means that SLR will also vary from region to region. This interactive map by Perrette and his team is a great demonstration of how SLR varies throughout the world under different emissions scenarios.

Sea level is expected to rise 25 cm higher in the Bay of Bengal than along the Dutch coast – in fact, SLR here is 10-20% higher than the global average! Why? It’s all to do with land ice and local gravity fields. Local gravitational effects suppress the effect of meltwater from the Greenland Ice Sheet along the Dutch coast, but there is no similar force keeping water back in the Bay of Bengal. In addition, land ice is responsible for a far greater portion of the SLR in the Bay of Bengal than it is along the Dutch Coast. This is no surprise, when you consider its close proximity to the Himalayas and other Asian mountains, but it adds an extra dose of uncertainty when predicting SLR here as land ice contributions to SLR remain a challenge to climate modellers.

Gravitational forces influence global, as well as local, water distribution. The declining mass of ice caps at the poles means there will be weaker gravitational forces acting on Arctic and Antarctic waters. The weakening of these poleward-pulling forces will cause water to be redistributed throughout the globe, resulting in greater SLR at the equator than at the poles. So, despite the large local variations in SLR, its effects will be felt most in the tropics and at the equator.

By Sara Mynott

References:

Perrette, M., Landerer, F., Riva, R., Frieler, K., and Meinshausen, M. (2013), A scaling approach to project regional sea level rise and its uncertainties, Earth Syst. Dynam., 4, 11-29, doi:10.5194/esd-4-11-2013.

Riva, R. E. M., J. L. Bamber, D. A. Lavallée, and B. Wouters (2010), Sea-level fingerprint of continental water and ice mass change from GRACE, Geophys. Res. Lett., 37, L19605, doi:10.1029/2010GL044770.