# Observation and modeling of platelet ice fabric in McMurdo Sound, Antarctica - **Authors:** D. E. Dempsey; P. J. Langhorne; N. J. Robinson; M. J. M. Williams; T. G. Haskell; R. D. Frew - **Journal:** Journal of Geophysical Research: Oceans, vol. 115(C1), article 2008JC005264 (2010) - **Publisher:** American Geophysical Union (AGU) - **DOI (version of record):** [https://doi.org/10.1029/2008jc005264](https://doi.org/10.1029/2008jc005264) - **This page:** https://ddempsey.github.io/papers/2010-dempsey-observation-modeling-platelet-ice-fabric/ - **Open PDF:** [https://ddempsey.github.io/papers/2010-dempsey-observation-modeling-platelet-ice-fabric/dempsey2010-observation-modeling-platelet.pdf](https://ddempsey.github.io/papers/2010-dempsey-observation-modeling-platelet-ice-fabric/dempsey2010-observation-modeling-platelet.pdf) (Published version) - **Full text (Markdown):** [https://ddempsey.github.io/papers/2010-dempsey-observation-modeling-platelet-ice-fabric/fulltext.md](https://ddempsey.github.io/papers/2010-dempsey-observation-modeling-platelet-ice-fabric/fulltext.md) - **Notice:** Published version, posted under AGU's author self-archiving policy. © 2010 American Geophysical Union. All Rights Reserved. Version of record: https://doi.org/10.1029/2008jc005264 - **Also in:** UC Research Repository, https://hdl.handle.net/10092/107328 - **Explainer:** all explainers on this site are AI-generated. - **BibTeX:** https://ddempsey.github.io/papers/2010-dempsey-observation-modeling-platelet-ice-fabric/citation.bib - **Topics:** Sea ice - **Keywords:** Sea ice; Geology; Ice crystals ## In brief Ice crystals that grow in supercooled water from beneath Antarctic ice shelves are frozen into sea ice. Using ice cores and a crystal growth model, we infer that western McMurdo Sound received a larger and more persistent supply of these crystals than the east in 2007. - In late 2007, loose ice platelets were recovered from ocean access holes at the three western sites on our transect, and none appeared at the three eastern sites. - We identify resumed columnar ice, a coarse crystal type beneath platelet ice, and suggest it forms once platelets stop arriving from the ocean. - In our model, platelets settling on a flat ice base produced a loose layer with no preferred crystal orientation within about 25 mm of ice growth. - In our model, a loose platelet layer forms when more than about 5 × 10^5 platelets arrive per square metre per day, with ice growing 10 mm per day. - Read as a qualitative indication of platelet supply, grain boundary density in the cores implies larger and more persistent platelet fluxes in the west, where the upper ocean was supercooled. ## The problem Around 44% of the Antarctic coastline has an ice shelf (Drewry et al., 1982). Melting at the base of an ice shelf releases cool, fresh water that rises. As the pressure falls, its freezing point rises. The water becomes supercooled, or colder than its freezing point (Foldvik and Kvinge, 1974). Ice crystals then grow in it. Theoretical estimates put the contribution of this process at up to 0.2 m of sea ice thickness over large areas of the Southern Ocean (Hellmer, 2004). In McMurdo Sound, these crystals arrive as thin discs called platelets. They accumulate beneath the landfast sea ice as a loose, porous subice platelet layer. As the sea ice thickens over winter, it freezes around the upper part of this layer to form incorporated platelet ice. Our goal in this study was to determine what can be inferred from the crystal structure of this ice about the conditions in which it formed. ## What we did Between 16 November and 1 December 2007, we drilled ice cores at six sites on an east–west line across McMurdo Sound, labelled W1 to W3 in the west and E1 to E3 in the east. We photographed thin sections of the ice between crossed polarising filters. These filters distinguish crystals by the direction of their c axis. From these images we measured grain boundary density, the length of boundary between crystals per unit area (Eicken et al., 1994). We also measured ocean temperature and salinity beneath the ice at each site. We then built a 3-D computer model that grows virtual platelets pixel by pixel (Ohashi et al., 2004). In the model, loose platelets rise, touch the ice and pivot until they are stable. They then grow in place until the advancing base of the sea ice freezes them in. ## What we found Columnar ice made up 44% of the eastern cores but about 5% of the western cores, where fine-grained and platelet ice dominated. Loose platelets were recovered from the holes drilled for ocean measurements at W1 to W3, and none at E1 to E3. The upper 20 m of the ocean was a few hundredths of a degree below its freezing point at W1 and W2. Elsewhere it was above freezing, by up to 0.2°C at E3. We also identify a crystal type that we call resumed columnar ice. It appears beneath platelet ice in the three eastern cores. We hypothesise that, once platelets stop arriving, it forms by geometric selection, in which crystals with near-horizontal c axes grow larger and the number of crystals falls. In our model, platelets left to grow without new arrivals reproduced the roughly linear increase in grain size with depth seen in natural sea ice. Our sensitivity runs suggest that the model would match the observed rate of increase with a larger growth anisotropy (sideways growth relative to thickening). We suggest that this ice in a core indicates the end of the supply of platelets from the ocean. In our model, the first platelets to reach a flat ice base settled with their flat faces horizontal. Later platelets tilted against them and the base became uneven. After about 25 mm of ice growth there was no preferred orientation, as in our cores. We could not check the cores for flat-lying first platelets. In the model, a loose layer builds up when platelets accumulate beneath the ice faster than the advancing ice base freezes them in. The critical platelet flux for this is proportional to the packing efficiency (the fraction of the layer occupied by ice) and to the rate of ice growth. For platelets 4 mm across, packing of 5% to 7% and ice growth of 10 mm per day, it is 4.5 to 6.3 × 10^5 platelets per square metre per day. By our arithmetic, that is 45 to 63 platelets per square centimetre each day. Because ice grows more slowly later in the season, we suggest that platelet layers are more likely to form then.
In the model, grain boundary density rose with platelet flux only at low fluxes. Provided the ice frozen between platelets contributes a constant share, grain boundary density is a qualitative indication of platelet supply. We interpret its increase with depth in cores W1 to W3 and E2 as a relative increase in platelet supply. We attribute its decrease with depth at E1 and E3 to the supply ending and geometric selection resuming. From these trends, we infer that western McMurdo Sound received larger and more persistent platelet fluxes than the east. This is consistent with the supercooling we measured. ## Limitations Winter breakouts of the sea ice meant that our sites froze over at different times. Our transect also does not represent the sea ice of the whole Sound. Because our model omits heat flow, it cannot predict salinity profiles or how the ice fraction of the layer changes. The modelled packing efficiency of 5% to 7% is lower than the 20% estimated for the natural layer in McMurdo Sound (Bunt and Lee, 1970). We attribute the difference to growth and movement of platelets within the layer that these simulations omit. With 20% packing, the same equation gives a critical flux three to four times higher (our arithmetic). Because it treats only mechanics, the model cannot distinguish draped from bladed platelet ice, whose occurrence may therefore depend on thermodynamic conditions. Including heat flow in the model would improve the accuracy and scope of its predictions. ## Glossary An **ice shelf** is a thick sheet of glacier ice that flows off the land and floats on the ocean. **Supercooled** water is liquid water colder than its freezing point. **Landfast sea ice** is sea ice that is attached to the coast and does not drift. A **platelet** is a thin, disc-shaped ice crystal that grows in supercooled seawater. The **subice platelet layer** is a loose, porous accumulation of platelets beneath the sea ice. **Incorporated platelet ice** is the part of the platelet layer that has been frozen into the sea ice above it. A **thin section** is a slice of ice thin enough for light to pass through. The **c axis** is the crystal direction perpendicular to the flat face of an ice platelet. **Columnar ice** is sea ice made of long, vertical crystals that grow downwards as seawater freezes onto the base of the ice. **Geometric selection** is the process in which, among crystals growing downwards together, those with c axes close to horizontal grow larger and the number of crystals falls. **Grain boundary density** is the length of boundary between crystals per unit area of a thin section. **Packing efficiency** is the fraction of the volume of a loose layer of crystals that is occupied by ice. **Draped** platelet ice has centimetre-sized crystals with curved boundaries, and **bladed** platelet ice has tall, blade-like crystals with sharp boundaries. A **platelet flux** is the number of platelets arriving beneath the ice per square metre per day. ## Abstract During the annual growth of landfast ice in McMurdo Sound, Antarctica, an episodic flux of platelet ice crystals from the ocean contributes to the build up of a porous subice platelet layer, which is steadily incorporated into the sea ice cover as it thickens over winter. In November 2007, we examined the spatial variability of these processes by collecting sea ice cores, with simultaneous oceanographic observations, along an east‐west transect in the sound. Previously identified draped and bladed platelet ice types were observed. In addition, we identify resumed columnar growth which appears to be a result of geometric selection from the subice platelet layer after the arrival of new platelet crystals from the ocean has ceased. A numerical model of mechanical platelet ice processes is developed that predicts crystal texture and c axis distributions, producing virtual incorporated platelet ice with known growth history. This model demonstrates how a disordered subice platelet layer arises from an initially flat interface and suggests that such a layer is more likely to form later in the growth season. The model also suggests how the grain boundary density in incorporated platelet ice responds to changes in the flux of loose platelet crystals from the ocean. Application of this result to our 2007 platelet ice observations indicates that sea ice in western McMurdo Sound is subject to larger and more persistent platelet fluxes than the ice in the east. This is consistent with the pattern of in situ supercooling just beneath the ocean surface. ## Cite Dempsey, D. E., Langhorne, P. J., Robinson, N. J., Williams, M. J. M., Haskell, T. G., Frew, R. D. (2010). Observation and modeling of platelet ice fabric in McMurdo Sound, Antarctica. Journal of Geophysical Research: Oceans, 115(C1), 2008JC005264. https://doi.org/10.1029/2008jc005264 ```bibtex @article{dempsey2010observation, title = {Observation and modeling of platelet ice fabric in McMurdo Sound, Antarctica}, author = {Dempsey, D. E. and Langhorne, P. J. and Robinson, N. J. and Williams, M. J. M. and Haskell, T. G. and Frew, R. D.}, journal = {Journal of Geophysical Research: Oceans}, year = {2010}, volume = {115}, number = {C1}, pages = {2008JC005264}, publisher = {American Geophysical Union (AGU)}, doi = {10.1029/2008jc005264}, url = {https://ddempsey.github.io/papers/2010-dempsey-observation-modeling-platelet-ice-fabric/} } ```