Dynamics of Large Pelagic Ice Crystals in an Antarctic Ice Shelf Water Plume Flowing Beneath Land-Fast Sea Ice

Observations of boundary-layer processes and ice crystal behaviour in an outflow region from the Ross/McMurdo Ice Shelves are presented. From a fast ice field camp, we captured the kinematics of free-floating relatively large (many 10s of mm in scale) ice crystals that were advecting as well as aggregating in a depositional layer on the sea ice underside (SIPL, sub-ice platelet layer). Simultaneously, we measured the background oceanic temperature, salinity, currents and turbulence 15 structure. At the camp location the total water depth was 536 m, with the uppermost 50 m being in-situ super-cooled. Tidal flow speeds had an amplitude of around 0.1 m s and the resulting under-ice boundary layer sustained turbulent dissipation rates as large as =10 W kg. Acoustic sampling (200 kHz) identified three classes of backscatter (1) large individual highly mobile targets, (2) echoes from large, individually identifiable suspended crystals and (3) a varying background, presumably of very 20 small (frazil) crystals. This second class of backscatter was associated with crystal sizes far larger than typical, certainly larger than anything normally described as frazil, and some individuals at least were depositing close to “fully grown”. Measurement indicated crystal scales of the range 30-80 mm. The existence and settlement of this scale of crystal has implications for understanding SIPL evolution and the processes controlling the fate of Ice Shelf Water. 25


Introduction
Regional variability in Antarctic sea ice is a major issue for climate prediction, challenging models (Ludescher et al., 2019) and confounding communication of key issues to stakeholders and decisionmakers. With anthropogenically-induced warming oceans penetrating farther south, increased ice shelf basal melting is expected (Rignot et al., 2013;Kusahara 2020). Relatively warm ocean water penetrates 30 the interior of ice shelf cavities and induces melting on the ice underside. One driver of sea ice variability is the feed-back effect of meltwater exiting major ice shelf cavities (e.g. Holland et al., 2007;Langhorne et al., 2015). The resulting water is at the local freezing point temperature, as dictated by pressure and salinity. This water mixes with the ambient ocean resulting in a fresher, cold seawater https://doi.org/10.5194/tc-2020-249 Preprint. Discussion started: 12 October 2020 c Author(s) 2020. CC BY 4.0 License. plume that seeks out the fastest upward flow path on the shelf underside subject to the Coriolis force 35 and basal slope (MacAyeal 1985;Jenkins and Bombusch 1995;Smedsrud and Jenkins 2005;Stevens et al., 2020). These plumes will grow with sustained melting and/or decay with re-freezing removing their thermal deficit. If the plume persists sufficiently to reach the ice shelf edge it flows out beneath the neighbouring sea ice margin (Fer et al., 2012;Langhorne et al., 2015). At this point the basal slope driver of flow ceases and the persistence of the supercool plume is controlled by initial buoyancy, 40 growth of new ice, topography and mixing (Hughes et al., 2014). This step represents a critical phase in the passage of basal melt water where the greatest changes in drivers and environment are all located.
This supercool water drives sea ice growth by absorbing heat into the stratified upper ocean and facilitates the generation and growth of ice crystals (Robinson et al., 2014;McPhee et al., 2016;Hoppmann et al., 2020). 45 There is evidence that in some settings this ice formation occurs as buoyant crystals in the water column (Hoppmann et al., 2015;2020;Frazer et al., 2020). If these crystals grow slowly, remaining sufficiently small that viscosity dominates, then they are mainly passively advected. Typically, this is the scale (<1 mm) at which crystals are thought to exist. However, if they grow sufficiently large whilst suspended, then buoyancy-driven thin disk mechanics must dominate their trajectory (Jordan et al., 50 2015). The extent of a plume has been modelled using schemes that develop and transport crystals, again with a focus on mm-scale crystals (e.g. Holland et al., 2007;Hughes et al., 2014), or if not, then in a "bed-load" and so essentially a part of the ice-ocean interface (Robinson et al., 2014). Sampling challenges make it difficult to build up spatial appreciation of the crystal metrics and growth rate in the SIPL but one correlation that emerges is that SIPL thickness and supercooled seawater are co-located 55 (Langhorne et al., 2015;Brett et al., 2020).
By following the ice shelf water plume as it evolves in space and time, it is possible to look at ice growth and thermal relief (e.g. Smedsrud and Jenkins 2005;Hughes et al., 2014). A recent review by Hewitt (2020) identifies issues like crystal growth, the role of sediments and the limited availability of observations as being key issues for the advancement of understanding at the ice shelf scale. At the 60 larger regional to global scale the challenges lie more with sea ice production and water mass formation as coupled models seek to combine the ocean, atmosphere and ice structure (e.g. Roach et al., 2018;Richter et al., 2020;Moorman et al., 2020). Uniformly these studies identify the need for more observations both at the process, and monitoring, scales.
A decade-long sequence of sea ice camps in the McMurdo Sound region (Robinson et al., 2020) 65 have revealed that these platelets form a coherent layer on the underside of sea ice (a sub-ice platelet layer, SIPL Hunkeler et al., 2015;Wongpan et al., 2015;Hoppmann et al., 2020) into which they are eventually incorporated (Smith et al., 2001;Langhorne et al., 2015). The McMurdo Ice Shelf, a small ice shelf that sits between the Ross Ice Shelf (the largest ice shelf on the planet by area) and McMurdo Sound. It includes a region called "the Dirty Ice" because of the substantial rock debris visible on the 70 https://doi.org/10.5194/tc-2020-249 Preprint. Discussion started: 12 October 2020 c Author(s) 2020. CC BY 4.0 License.
shelf surface and has been described as "perhaps the strangest ice shelf in the world" (Debenham, 1965;Atkins and Dunbar 2009). This material is partly marine in origin, as sediment entrained into the growing marine SIPL on the shelf-underside finds its way to the surface (Campbell and Claridge 2003).
This, in itself, is evidence of supercool oceanic conditions. Hoppmann et al. (2020) review our present understanding of Antarctic platelet ice and makes it 75 clear the topic is still in a discovery phasepartly due to the challenges of making comprehensive observations. At the same time, modelling approaches have needed to advancecreating a tension. It is likely that the deposition and formation of ice crystals at a range of scales influences interfacial momentum transfer, sea ice composition and strength as well as ecological habitat throughout localised parts of Antarctic coastal waters. While geophysical boundary-layers are well understood, a number of 80 questions arise around unique aspects of the present situation and provide a focus for the present study.
(1) Is there evidence of large pelagic crystals? (2) Is there a relationship between crystal behaviour and the turbulent under-ice boundary-layer structure? (3) Does sediment from the Dirty Ice influence the mechanics? (4) What are the large-scale implications of such finescale mechanics?

Location and Camp
The "K131 sea ice camp" (Antarctica New Zealand logistics event designation, Stevens et al., 2018) (Langhorne et al., 2015), while avoiding too many platelets (i.e. SIPL too deep to easily penetrate a conductivity temperature depth -CTD -profiler through) and also avoiding the substantial tidal currents encountered further towards Haskell Strait (Figure 1a, Stevens et al., 2009;Mahoney et al., 2011). The K131 camp consisted of modified shipping containers with cut-out floors allowing 95 access to the ice and ocean below. A hot-water cutter was used to melt through and remove the sea ice in blocks. It was notable that upon removal of the blocks, the water which filled the hole appeared milky but that this gradually dissipated over the subsequent days. After 12 days of operations and many seal occupations of the holes, the hole water was fully flushed and very clear. We speculate that the water in the hole was initially from the melting of the sea ice and upwards drainage from the SIPL and 100 contained sufficient levels of sediment to be visible but that over time this was replaced with clear ocean water. With regard to the regional sampling context, the present data were collected south of where the later Frazer et al. (2020) acoustic sampling took place. Furthermore, the analysis of under ice https://doi.org/10.5194/tc-2020-249 Preprint. Discussion started: 12 October 2020 c Author(s) 2020. CC BY 4.0 License. boundary roughness data synthesis described in Robinson et al. (2017) include some data from the same field camp as here, but with instruments focused on the ice underside. suggests that the SIPL is not always a tightly interlocked matrix of crystals. The dimensions of each 115 crystal were then quantified by manual measurement (done immediately to avoid significant melting or growth). It is possible that these represent a distinctly different set of sizes to the suspended pelagic crystals, but large crystals were definitely observed in video rising. It is likely however that this approach is biased towards larger crystals (> 5 mm diameter). Crystal size data from a nearby camp from the following year (77.8183°S 165.4059°E, November 2016see Robinson et al., 2020) are also 120 included.

Video and echosounder records
While qualitatively very useful, video/visual observations are challenging to interpret quantitatively because of light variability, lens distortion and reflection/viewing angle effects.
Additionally, it was difficult to see individual crystals arrive and settle as the light would saturate the 125 camera. Penrose et al. (1994) describe how acoustics provide a more consistent picture of crystal behaviour. We used a Simrad EK60 200 kHz echo sounder recording acoustic backscatter at 1 Hz with 4 cm vertical resolution over a sampling cone that is 7 degrees across so that at a depth of 25 m the cone is three m wide. The cone has side-lobes in the upper 5 m of the water column that are as wide as 30 deg. The beam-width is not particularly critical so long as it is wide-enough that scatterers register

Profiling instruments
Water column temperature and salinity were recorded using a SBE19+ CTD profiler that was calibrated pre-and post-experiment resulting in an accuracy of ~5 mK. Care was taken to avoid ice growth on the CTD sensors, and also to thermally equilibrate prior to profiling (Robinson et al., 2020). The example 145 used here is a down-cast from an instrument that was held for one hour at 200 m prior to bringing up into the hole very briefly and then profiling downwards.
In addition to the CTD profiler, a Rockland VMP 250 microstructure profiler was deployed to quantify turbulent mixing by capturing fine and microstructure scale variability. This loose-tethered profiler falls at ~0.65 m s -1 , recording velocity shear and temperature at the microscale and fine-scale 150 temperature and conductivity along with some other properties. The microscale shear enables estimates of energy dissipation rate  (Wolk et al., 2002) and has previously been successfully deployed beneath Antarctic fast ice (Robertson et al., 1995;Stevens et al., 2009;Fer et al., 2012). Eighty-one microstructure profiles were recorded over a three-day period (1-3 Nov. 2015).

Moored instruments
155 Three pairs of current meters (Aanderra RCM 11) and conductivity-temperature-pressure sensors (Seabird Electronics SBE 37), both sampling at two-minute intervals, were deployed on a suspended mooring located at 77º 51.903' S 166º 00.351' E, 200 m north east of the main sampling location. The instrument pairs were deployed at 32, 82 and 374 m beneath the surface. In addition, ten SBE 56 temperature loggers sampling at 1 Hz were deployed at 5, 10,15,20,43,53,141,199,257,315 m. All 160 the upper instruments were affected by icing issues (Robinson et al., 2020) to varying extents and only used in specific instances here.

Background water column conditions
Water column temperature structure showed that the ocean temperatures were mostly above the However, they are not the only advective process, as a period of flow near the start of the experiment showed speeds in excess of 0.1 m s -1 and consistently moving towards the north (i.e. away from the ice 180 shelf). Temperature and salinity were consistent over the period but still responding to the tide at times.

Ice crystals and backscatter
The majority of the measured crystal dimensions ranged from 5 mm through to 200 mm ( Figure   4). The average plan-view dimension in 2015 was 93 mm (with a slightly larger 101 mm equivalent measured the following year). Thicknesses were 2-10 mm, with the thicker ones clearly multi-layered 185 (thickness was not measured in the subsequent data set). The ice crystals ( Figure 4a) were often larger than 100 mm in apparent scale ( Figure 4c) and while these are found in the well-defined SIPL beneath the sea ice, it became clear that there was a constant supply of crystals from depth, some of which were already of large scale.
There appeared to be two types of behaviour from visual video observations ( Figure 5). The 190 first was seen at around 10 m depth platelet crystals were being advected horizontally but with some randomness to direction and not always with an obvious upwards component. The second type was seen in imagery from just beneath the ice (~ 1 m) which showed a more ordered region of suspended crystals, especially at slack water. The smaller ones drifted slowly horizontally coherently while the larger individuals were occasionally and independently seen rising into the SIPL. 195 The 200 kHz echosounder provided a new perspective on the presence and behaviour of these suspended crystals as various acoustic backscatter conditions were observed over the nine days of sampling ( Figure 6). These conditions included large individual biological agents observed against a slowly varying background (Figure 6a) on occasionally with a more rapidly varying background ( Figure   6b). It was also common to observe a varying intensity in the background field ( Figure 6c), but 200 individual target streaks would persist through the signal variation. The target strength was not a reliable separator however, as there would occasionally be strong scatterers that simply rose and entered the SIPL. Interpretation of video suggests these are relatively large crystals.
The analysis of crystal rise speed (Figure 7) found only a moderate bias to upwards flows and quite small velocities (mostly less than 1 cm s -1 ). With the larger crystals being around 7 cm in 205 diameter and rise speeds of the order of 1 cm s -1 , this implies a Reynolds number Re=0.01x0.07/10 -6 =700. Instability behind a buoyant disk commences well below this at Re=~60 (Natarajan and Acrivos, 1993). However, a significant proportion of crystal speeds are downwardsi.e. against buoyancy (positive velocity as shown in Figure 6b). This suggests that the upper water column is some combination of (1) internal wave motion that is as likely to be downward as it is 210 upwards, (2) isotropic shear-driven mixing and/or (3)  variations in background signal amplitude seen in Figure 6b and c occur around 10 s intervals so around 240 9000 cpd. These sit well into the high frequency tail of the spectrum (Figure 8). Another scale of variability that will be apparent at least close to the surface is the SIPL underside has around a 2-5 m undulation which coupled with a 0.1 m s -1 implies a 20 s variation. This too falls to the right in the high frequency content.

245
The VMP profiles revealed mostly good quality turbulent spectra (Figure 9) allowing for reliable estimation of  which provides insight into the dynamics of the vertical structure of suspended crystals The nature of the ISW outflow is not consistent from year to year. In the present context they primarily captured the background field, as the majority of their crystals were estimated to be around 1 mm in 285 scale and any large crystals would have been removed by the processing. Video observations and target behaviour support the contention that the majority of rare very "bright" individual targets were large faunaranging from fish, fish schools through to seals. However, it was far more common to observe a hybrid of the 2 nd and 3 rd conditions whereby the sampled field consisted of many less bright but still clearly individual signals against a coherent background field. The present situation is downstream of 290 the Ross Ice Shelf cavity within which residence times in zero light likely to be in the range of 1-5 years (Reddy et al., 2010;Stevens et al., 2020). This reinforces the contention that the targets and continuum are ice related rather than suspended biological or sediment.
Here we refer to the large, mature crystals at depth as pelagic crystals to distinguish them from crystals already integrated and growing in the SIPL. Laboratory and numerical work demonstrate that 295 buoyant disks rotate so that their flat face is roughly horizontal although in some circumstances it is possible for an oblique equilibrium to exist which might be the cause of some of the horizontal motion observed here (Fabre et al., 2012). Furthermore, the behaviour of rising disks can be connected back to the initial conditions, suggesting attention be paid to the spontaneous growth from a very small nucleus (Daly 1984;Tchoufag et al., 2014 where the one-to-one equivalence has no dynamic significance). Rees Jones and Wells (2018) use one of the faster rise rates from the same set of results which is applicable to smaller crystal sizes (e.g. none of which conforms to the drag based estimates described in Daly, 1984) whereas Matsumura and 305 Oshima (2015) use a fixed rise rate of 1 mm s -1 , again for small crystals.
Production of crystals at depth and its subsequent integration into sea ice is a key step in the formation of the SIPL, at least at this location (Hoppmann et al., 2015;Hunkeler et al., 2015). This has https://doi.org/10.5194/tc-2020-249 Preprint. Discussion started: 12 October 2020 c Author(s) 2020. CC BY 4.0 License.
several implications, the most important being an alternate pathway for platelet arrival and structuring of the SIPL. If they arrive essentially mature at the ~5 cm scale, this is very different to arriving at the 1 310 mm scale and then growing (Dempsey et al., 2010) and the categories of backscatter described earlier suggest both happen. The issue of orientation suggests that the arrival velocity will be slower than if they were to rise in some other orientation with a reduced drag profile. Thus, there may be a correlation between rise speed and packing in the SIPL. Potentially this alignment relates to the large-scale variability seen in Figure 6c whereby large parts of the domain change backscatter but individual 315 scatterers are clearly seen through the transition. In other words, horizontal flow is slow, yet the scattering still changes as crystals re-orientate themselves within the sensing volume.
The influence of pelagic crystals can be represented in larger-scale models such as Kim et al. (2006) and Roach et al. (2018). However, it will require some local-scale mechanics. For example, Dempsey et al. (2010) and Wongpan et al. (2015) simulate this aggregation by injecting a continuous 320 flux of crystals from the ocean at the same time as recording (i) the rate at which the sea ice incorporation front moves downward and (ii) below this front but still within the SIPL, crystals continue to grow. In their approach, they maintain different size classes for the deposited systems as opposed to those floating up from below. This implies that there is an ability for the rising crystals to fill in the interstices of the SIPL, reducing the void fraction. This is a point made by Dempsey et al. (2010) who 325 quantified the flux rate of 4 mm diameter platelets required to grow the observed sea ice to be of order 10 6 crystals m -2 d -1 .
The pelagic growth to the crystals also means that the brine rejection will happen essentially within the upper ocean layer as well at the sea ice underside. This affects the upper layer turbulence and entrainment. In turn, this influences the persistence and fate of the ISW plume. This shifts the 330 buoyancy flux inherent in the energy conversion from a boundary process to what is effectively an "internal buoyancy source".

Crystal behaviour in the turbulent under-ice boundary-layer
It is useful to compare vertical rise rate of crystals wc with turbulent mixing in the water column.
The present profiler-resolved dissipation rates are comparable with Fer and Widell's (2007) data from 335 beneath Arctic sea ice in a fjord. However, while both studies observed turbulent energy dissipation rates in the range =10 -7 to 10 -6 W kg -1 , their results were from a faster-moving water column. It would appear the rougher ice underside here increases the turbulence to provide apparent matching conditions.
In the upper water column (5-10 m), the dissipation rate is an order of magnitude, or more, greater than the value expected if we match the deeper . One hypothesis is that we are seeing brine rejection and 340 associated enhanced turbulence. Regardless, the  provides a dynamic context for considering how the pelagic platelet crystals behave. The u* is a combination of convection-induced turbulence and draginduced stirring. Observations of boundary-layers beneath platelet ice have suggested that the drag https://doi.org/10.5194/tc-2020-249 Preprint. Discussion started: 12 October 2020 c Author(s) 2020. CC BY 4.0 License.
coefficient is a factor of 6-30 times larger than might be expected for a smooth, melting ice surface (Robinson et al., 2017). This reinforces the apparent paradox that melting ice (the shelf basal 345 underside) produces re-freezing that then affects how the entire system circulates. A key knowledge gap highlighted then is the under-shelf mixing in the basal melt layer. This will influence not only the sub-shelf re-freezing but the amount of supercooled water being ejected into the sea ice system. The apparent bi-directional variability in vertical motion (Figure 7) implies wave-type motion. 375 However, the density structure gives no indication of stratification in the upper 30-50 m that would support internal waves. It is possible that slow horizontal advection of convective processes moving down and upwelling flows past the observation location might also be a factor influencing the measured quantities.

380
While suspended sediment has been implicated in atmospheric ice formation (Kulkarni et al., 2014), it has not been identified as a nucleation point for marine ice crystal growth. Daly (1984) indicated that the thermodynamics precludes sediment playing a lead role in nucleation. However, there may be other pathways i.e. through aiding aggregation of ice crystals that can support nucleation.
Previous observations in the region (Robinson et al., 2017) had not provided strong evidence that The coincidence of a substantial sediment load suggests that, in this location at least, this might enhance crystal formation. However, a ready source of sediment is potentially a rare phenomenon at the continent-wide shelf edge suggesting that it is not a major controller of crystal production. It remains highly likely at grounding line zones where a confluence of ice shelf water outflow, glacially-driven sediment supply, tidal and subglacial resuspension and supercooling-induced re-freezing all likely 400 combine. Furthermore, substantial platelet ice layers exist in many ice shelf influenced locations where there is no obvious sediment source (Langhorne et al., 2015;Hoppmann et al., 2020). The intriguing speculation that follows from this iswhat role might sediment play in crystal nucleation near ice shelf grounding-lines and would we be able to tell if this influences marine ice formation (e.g. Fricker et al., 2001)? 405

What are the large-scale implications of flow-crystal interaction?
Local datasets such as the present experiment need to be placed into regional and continent-wide perspectives. There is a clear regional bias due to the majority of data coming from only a few field locations (Hoppmann et al., 2020). Debenham's comment about the region being the "strangest on the planet" suggests there may be some unique features that can't be generalised. While he wrote this in shows a striking similarity between SIPL and the location of the Dirty Ice. However, this also corresponds to where one would expect basal cavity outflow so supply of ISW is available (Hughes et 420 al., 2014). Following this water north, a number of studies have concluded that the ice shelf water from the Ross/McMurdo system persists for a hundred or more km northward (Hughes et al., 2014;Robinson et al., 2014). Richter et al. (2020) state that ice shelf-sea ice-ocean connections remain the major outstanding challenge in models operating at the continental and global scale. Injection of the range of processes 425 described here into modelling approaches that typically resolve scales around 2 km will be a challenge.
A starting point might be the role present large platelets play in the McMurdo Sound polynya formation (Dai et al., 2020). Polynya processes in particular become critical as they represent spatially-constrained phenomena driven by short high energy wind events but pre-conditioned by ice shelf water, that then drive formation of new sea ice. It is the by-product of this sequence that generates high salinity shelf 430 water that ultimately has a global thermohaline impact. Modelling of such processes thus may potentially need to better account for the nature of ice shelf water plumes.
The topic is clearly still in a discovery phase with many fundamental questions remaining unanswered. This work suggests research themes for understanding sea ice formation near ice shelves should focus on the role of convection driven by SIPL crystal growth in modifying the turbulence in the 435 upper water column and the feedbacks to the turbulence. In addition, the possible links between availability of nucleating material, crystal production and fate need to be examined, especially as to how this might support the formation of large, suspended ice crystals.  https://doi.org/10.5194/tc-2020-249 Preprint. Discussion started: 12 October 2020 c Author(s) 2020. CC BY 4.0 License.