The Inferred Formation of a Subice Platelet Layer Below the Multiyear Landfast Sea Ice in the Wandel Sea (NE Greenland) Induced by Meltwater Drainage

Abstract

Oceanographic and ice‐mass‐balance records are presented from two moorings deployed on landfast multiyear ice in the Wandel Sea (North Greenland) during June–August 2015. Here we show that the melting and drainage of >1 m of snow from June 14 to July 14 created a double‐diffusive vertical stratification which resulted in supercooling of water and enabled the formation of platelet crystals below the sea ice. Although the effect of supercooling, with temperatures up to 0.5°C below the freezing point, might be overestimated considerably in our records, this process led to the formation of ∼1.1–1.2 m‐thick subice platelet layer. While warm water temperatures lead to the complete loss of this layer at one mooring site, the layer persisted through summer and became incorporated into the congelation ice at the second site. The warm water that melted out the platelet layer can be ascribed to two different sources: (1) in situ heating from solar radiation resulting in a temperature increase up to 0.8°C in late July and (2) advection of warm surface water (with temperatures up to 3–4°C) from the ice‐free coastal regions in mid‐August. The combination of processes causing the seasonal growth of a platelet layer and either its subsequent ablation or incorporation into congelation ice is discussed with respect to the ice‐mass balance and stability of the landfast ice cover in the Wandel Sea. Furthermore, this study provides evidence for the formation of a platelet layer in the Arctic, a phenomenon that historically has only been observed in the Antarctic.

Variability of the Pacific‐Derived Arctic Water Over the Southeastern Wandel Sea Shelf (Northeast Greenland) in 2015–2016

“…This was accompanied by significant modifications through the entire water column (Figures b–d). The under‐ice 10‐m thick water layer shows warming (up to ~2.5 °C) and freshening (down to 17–20) attributed first to the drainage of snow meltwater through the landfast ice (JD196–216) and then to a pulse of warm meltwater advected from the ice‐free area generated over the western flank of the glacier trough to southwest of the mooring location (JD226–247, Figure )—Kirillov et al (). This is confirmed by the velocity time series at 8 m depth showing consistent northeastward flow starting JD218 (Figures a and a).…”

“…The Pacific‐derived halostad in the Wandel Sea is characterized by weak vertical salinity difference (salinity 30 to 31.2–31.5) and water near the freezing point (Dmitrenko et al, ). At the top of the Pacific‐derived halostad, the subsurface halocline layer with a strong vertical salinity gradient (salinity 1 m −1 ) separates the halostad from the relatively fresh (salinity of 16–21, not shown) surface layer that forms locally from snow and sea‐ice meltwater (Kirillov et al, ) and freshwater from the glacial runoff (Bendtsen et al, ). Below the Pacific‐derived halostad, the Atlantic‐derived halocline can be found with salinities between 31.5 and 34.…”

“…We suggest that the seasonality in water column properties in the southeast Wandel Sea below the under‐ice meltwater layer is generated by the onshore lateral displacement of the coastal branch of the PW outflow from the Arctic Ocean to the western Fram Strait driven by the annual cycle in wind forcing observed over the Wandel Sea continental slope. Seasonality of the solar radiation has little influence on the intermediate water layer, which is isolated year‐round from the surface by the subsurface halocline with exceptionally strong salinity (density) stratification (Bendtsen et al, ; Dmitrenko et al, ; Kirillov et al, ). Further indication of its isolation from the surface water is that during summer the water layer below 15 m depth exhibits salinification, while freshening is expected due to snow, sea‐ice, and glacier melt.…”

“…For the downward‐looking ADCP the first measured bin was located at 8 m depth (i.e., about 5 m below the ice), and ADCP measured the water layer down to 60–80 m depth. This ADCP stopped working on 26 December 2015, and there was a data gap from 7 July to 7 August 2015 when the ADCP transducer was blocked by a platelet ice layer (for more details see Kirillov et al, ). For the upward‐looking ADCP the first bin was at l70 m depth, and the ADCP sampled the water column up to 90–110 m depth.…”

“…The RDI ADCPs precision and resolution are ±0.5% and ±0.1 cm/s, respectively. The compass accuracy was estimated to ±8° due to the small horizontal component of the Earth's magnetic field in the Wandel Sea (for more details see Kirillov et al, ). The measured current direction was corrected for the local magnetic deviation (~18°W).…”