英文摘要: | Antarctica’s contribution to global sea-level rise has recently been increasing1. Whether its ice discharge will become unstable and decouple from anthropogenic forcing2, 3, 4 or increase linearly with the warming of the surrounding ocean is of fundamental importance5. Under unabated greenhouse-gas emissions, ocean models indicate an abrupt intrusion of warm circumpolar deep water into the cavity below West Antarctica’s Filchner–Ronne ice shelf within the next two centuries6, 7. The ice basin’s retrograde bed slope would allow for an unstable ice-sheet retreat8, but the buttressing of the large ice shelf and the narrow glacier troughs tend to inhibit such instability9, 10, 11. It is unclear whether future ice loss will be dominated by ice instability or anthropogenic forcing. Here we show in regional and continental-scale ice-sheet simulations, which are capable of resolving unstable grounding-line retreat, that the sea-level response of the Filchner–Ronne ice basin is not dominated by ice instability and follows the strength of the forcing quasi-linearly. We find that the ice loss reduces after each pulse of projected warm water intrusion. The long-term sea-level contribution is approximately proportional to the total shelf-ice melt. Although the local instabilities might dominate the ice loss for weak oceanic warming12, we find that the upper limit of ice discharge from the region is determined by the forcing and not by the marine ice-sheet instability.
Sea-level rise poses a future challenge to coastal regions worldwide13 and affects livelihoods and ecosystems in the most vulnerable regions already today14. Despite significant advances in ice-sheet modelling3, 4, 15, 16, 17, the largest uncertainty in future projections arises from the dynamics of the Antarctic ice sheet5. Ice shelves, the floating extensions of the ice sheet, modulate the ice-sheet flow through their back stress on the upstream glaciers, which is termed buttressing9. Although diminished shelves do not directly contribute to sea-level rise, the associated loss in buttressing can generate sea-level-relevant ice loss through the acceleration and thinning of upstream glaciers. In addition to difficulties with the boundary conditions and the bed and ice rheology, the quantification of the future Antarctic ice flow is complicated through ice instability and potentially abrupt changes in ocean heat transport. First, marine-based ice sheets with retrograde bed slope can experience unstable grounding-line retreat2. It is, for example, hypothesized that the Amundsen sector has entered a state of unstable retreat triggered by increased melting beneath the ice shelves of Pine Island and Thwaites Glaciers3, 4, 18. Ice shelves play an important role in modulating the instability through their buttressing effect, with strong buttressing being able to inhibit unstable grounding-line retreat10, 11. Second, atmospheric changes may trigger the breakdown of the Antarctic slope front in the ocean off the Antarctic coast. At present, most Antarctic ice shelves are surrounded by cold water masses near the freezing point. That is because the Antarctic slope front19 acts as a barrier for heat and salt exchange with the northern warmer and saltier water masses. Projections of the breakdown of this front in ocean simulations under atmospheric warming6, 7, 20 and under the southward shift of the Southern Westerlies21 raise concerns that increased ocean heat transport towards the ice shelves will boost future ice loss from Antarctica. It is of fundamental importance for the projection of future sea-level rise to understand whether global warming can trigger abrupt ocean warming beneath the ice shelves and induce instability of marine-based ice sheets. The tributary glaciers that feed the Filchner–Ronne ice shelf (FRIS, Fig. 1) rest on retrograde bed slope22 that suggests that unstable ice retreat is possible8. High-resolution ice-sheet simulations show that some of the glaciers can undergo accelerating retreat when sub-shelf melting is sustained at increased levels12. However, the FRIS glaciers are highly buttressed by the large ice shelf, which can suppress unstable retreat9, 10, 11. A reduction in size and volume of the FRIS can be expected under the abrupt intrusion of warm water into the ice-shelf cavity that is indicated by ocean models6, 7. It is unclear whether the strongly increased ice-shelf mass loss by the abrupt intrusion of warm water can induce the unstable retreat of the FRIS glaciers in a way that it will dominate the region’s contribution to sea-level rise. The large buttressing ice shelf suggests a tight link between ice-shelf melting and sea-level-relevant ice-sheet mass loss. The FRIS glaciers may therefore respond differently to ice-shelf melting than the ice-sheet-instability-dominated Amundsen glaciers3, 4.
| http://www.nature.com/nclimate/journal/v6/n1/full/nclimate2808.html
|