Sea ice in the Arctic Transpolar Drift in 2020/21: thermodynamic evolution of different ice types

Abstract. Sea ice properties are extremely inhomogeneous, in particular on the floe-scale. Different characteristic local features, such as melt ponds and pressure ridges, profoundly impact the thermodynamic evolution of the ice pack even in a kilometre-scale domain, and the associated processes are still not well represented in current climate models. To better characterize the freezing and melting of different types of sea ice, we deployed four sea ice mass balance buoys on an ice floe close to the North Pole during the second drift of the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) in August 2020. The study sites included level first-year ice, an open melt pond, and an unconsolidated ridge. The floe slowly drifted southwards from October 2020 to early March 2021 but shifted to a more rapid drift from March to July 2021. This drifting pattern, together with a large snow accumulation, relatively warm air temperatures, and a rapid increase in oceanic heat close to Fram Strait, determined the seasonal evolution of the ice mass balance. Storms, accompanied by higher air temperatures and enhanced ice dynamics, were the main cause of the formation of snow ice or superimposed ice. Although the 0.24-m deep melt pond was completely refrozen by 5 September, the relatively large snow accumulation and the heat storage with the rotten ice layer delayed ice basal growth beyond the last observation at this site in mid-February 2021. At the ridge site, the macroporosity of the unconsolidated layer was estimated between 0.005 and 0.755. The freezing of internal voids also delayed the ridge basal growth, which was not observed until 26 April 2021. Thus, the refreezing of ponded ice and voids within the unconsolidated ridges amplifies the anisotropy of the heat exchange between the ice and the lower atmosphere/upper ocean. Our results provide an important physical background for further interdisciplinary studies related to the MOSAiC observations and can be used to optimize the parameterization of freezing processes related to melt ponds and ice ridges in sea ice numerical models.