The thinning of floating ice shelves around Antarctica enhances upstream ice flow, contributing to sea-level rise. Ice-shelf thinning is now shown to influence glacial movement over much larger distances than previously thought.
At the beginning of July 2017, a giant iceberg — ten times the size of Manhattan — detached from the Larsen C Ice Shelf on the eastern Antarctic Peninsula1. Given that Antarctica stores enough freshwater to raise global sea levels by 60 m (ref. 2), and that ice shelves control ice fluxes into the Southern Ocean, this spectacular event reinforces anxieties surrounding the future stability of Antarctica in the context of global warming: what impacts will the Larsen C calving, and other similar events, have on the mass balance of Antarctica? Now, writing in Nature Climate Change, Ronja Reese and colleagues3 develop a model framework to quantify the current health of Antarctic ice shelves in light of idealized melt. This allows an estimation of the relative influence of each part of the ice shelf on the overall flow of the Antarctic Ice Sheet.
Floating ice shelves play an important mechanical role in controlling the movement of grounded tributary glaciers, providing a buttress — or barrier — that constrains ice flow into the ocean. If these ice shelves are weakened by a reduction in their size or thickness, upstream flow can be drastically accelerated. Following the disintegration of the Larsen B Ice Shelf in 2002, for example, the neighboring Hektoria, Green and Evans glaciers experienced a marked acceleration4, increasing annual ice fluxes from 2.7 km3 to 23.5 km3. Today, more than a decade after the Larsen B collapse, these glaciers are still losing mass at an increased rate. Similar features were also observed following the collapse of the Larsen A and Wilkins Ice Shelves in January 1995 and March 1998 (ref. 5), respectively.
While the catastrophic disintegration of ice shelves clearly exemplifies the instantaneous loss of the buttress effect and associated acceleration of flow, ice-shelf strength is also altered by various processes related to a warming climate. For instance, at the ocean–ice interface, intense melt driven by the onshore flow of circumpolar deep water has modified ice-shelf thicknesses all around Antarctica. Indeed, alongside iceberg calving at the ice shelf edge, this mechanism has been shown to be a dominant ablation process6. Furthermore, surface melt-water accumulation from warm winds has also been found to enhance the risk of ice-shelf collapse in East Antarctica7.
Although our process-level understanding has advanced considerably in recent years, the impact of ice shelves on glacier flow is strongly dependent on the location of ice-shelf mass loss. However, a spatially explicit quantification of this dependency was lacking. Now, using a simple modelling framework, Reese et al. quantify the impact of idealized ice-shelf thinning on the instantaneous flow of upstream grounded glaciers; they do this by removing a 1-metre-thick volume of ice over a 20 km2 area from each location, and repeating for many places on all ice shelves around Antarctica.
Reese et al. find high spatial variability in the capacity of ice shelves to buttress upstream-grounded glaciers. In general, removing ice close to grounding lines of fast-flowing glaciers has stronger effects than at the edge of the ice shelf. For example, the Filchner–Ronne and Ross Ice Shelves have strong buttressing effects across almost the entire grounding line; this is particularly worrying as it is also the location of the highest basal and surface melts4,7, threatening future stability. For Pine Island glacier, the largest contributor to sea-level rise in Antarctica, almost all its ice shelf plays an important buttressing role.
Interestingly, the method proposed by Reese et al. allows quantification of the distances over which ice removal influences fluxes at the grounding line. While in general the melt perturbations close to the grounding line have a local effect, the method allows detection of locations where tele-buttressing is at play. Again, for the Filchner–Ronne and Ross Ice Shelves, several critical areas can be located for which the local perturbation affects very remote grounding lines, up to 1,000 km away. It is of major concern that these critical areas are located close to the edge of ice shelves at places where ice is regularly lost by calving.
While clearly identifying locations vulnerable to ice-shelf thinning, these model results also reveal places with a very limited effect on upstream glacier flow; these ‘passive’ ice shelf regions8 could be removed without inducing any change in ice flux at the grounding line. While the calving of the Larsen C iceberg produced significant media attention due to the potential implications on the stability of the ice shelf, Reese et al. confirm the calving occurred in a passive area. Consequently, this spectacular calving event should have no instantaneous effect on upstream ice flow, and thus sea level. Similarly, unconfined ice shelves, such as the last portion of the Mertz ice tongue, are also identified as being passive.
Though a step forward in our understanding of ice-shelf impacts of Antarctic stability, the study by Reese et al. only provides information on the instantaneous response following a local ice loss. To fully quantify the future impact of ice-shelf loss to glacial flow, dynamical effects also need to be considered. This, however, is an extremely challenging task given multiple uncertainties, and non-linear feedbacks and processes9,10, including, for example, the link between hydrology and basal friction11.
Nevertheless, Reese et al. confirm the protective role of ice shelves in controlling Antarctic ice flow, but further highlight their increasing weakness under anthropogenic warming, and thus the potential heightened contribution of Antarctica to future sea-level rise. Moreover, while recent events at Larsen C may have occurred in passive regions, it remains to be seen whether future ice loss sets it on the same path as its disintegrated neighbours.
Hogg, A. E. & Gudmundsson, G. H. Nat. Clim. Change 7, 540–542 (2017).
Fretwell, P. et al. Cryosphere http://doi.org/f22853 (2013).
Reese, R. et al. Nat. Clim. Change https://doi.org/10.1038/S41558-017-0020-x (2017).
Rignot, E. et al. Geophys. Res. Lett. 31, L18401 (2004).
Scambos, T. A., Bohlander, J. A., Shuman, C. A. & Skvarca, P. Geophys. Res. Lett. 31, L18402 (2004).
Rignot, E. et al. Science 34, 6143 (2013).
Lenaerts, J. T. M. et al. Nat. Clim. Change 7, 58–62 (2017).
Fürst, J. J. et al. Nat. Clim. Change 6, 479–482 (2016).
Ritz, C. et al. Nature 528, 115 (2015).
DeConto, R. M. & Pollard, D. Nature 531, 591–597 (2016).
Pattyn, F. et al. Current Clim. Change Rep. 3.3, 174–184 (2017).
About this article
Cite this article
Gagliardini, O. The health of Antarctic ice shelves. Nature Clim Change 8, 15–16 (2018). https://doi.org/10.1038/s41558-017-0037-1
Antarctic ecosystem responses following ice‐shelf collapse and iceberg calving: Science review and future research
WIREs Climate Change (2020)
Scientific Reports (2019)