DynAMIC (Detecting episodes of Arctic sea ice Mass Imbalance)

The Arctic Ocean's sea ice cover has shrunk 75% in volume since the 1980s. This reduction hasn't happened gradually but in rapid periods of sea ice loss lasting between one and a few years. Years of rapid summer ice loss are preconditioned by the state of the ice cover in the preceding winter, and the state of the ice depends on winter storms that pass through the Arctic. Storms bring episodes of sea ice mass imbalance that violently reshape the ice cover. In DynAMIC we bring together different satellite sensors from ESA, EU-Copernicus and NASA to observe the 3D evolution of the sea ice directly, as storms reshape the ice cover in winter. We use radar sensors to image the floes (patches) of sea ice deforming, then we use altimetry sensors to measure the floes thinning or thickening as they deform. Such dynamic events have so far only be observed a few times from planes and ships, rather than from space, meaning that we don't really know how sea ice typically reacts to storms. Armed with a database of many sea ice "imbalance" events detected by satellites, we will be able to add realistic sea ice ridging to an advanced Norwegian model for the ice dynamics: neXtSIM. Model experiments will then, finally, show how local episodes of sea ice imbalance shape the Arctic ice cover for years of rapid summer ice loss. It is urgent that we create such tools for understanding climate changes in a rapidly-warming Arctic.

As the Arctic Ocean’s sea ice cover has thinned and weakened over the past three decades, it has become increasingly susceptible to episodes of sea ice mass imbalance that reshape the winter ice thickness distribution. It is unclear whether these dynamic events have contributed to record low Arctic sea ice areas in recent years because currently available satellite technologies only allow us to observe the rapid 3D (thickness + area) changes in 2D (area only). Moreover, we cannot yet model these events because the parameterizations for sea ice mass redistribution by pressure ridging do not capture the observed evolution of the ice thickness distribution over winter. The prospective EU polar-orbiting CRISTAL satellite mission will observe the 3D evolution of sea ice and its snow load directly, offering the perfect dataset to constrain ridging events in numerical models. But what can we do before CRISTAL is launched later in the 2020s? The DynAMIC project will combine existing satellites from ESA, EU-Copernicus and NASA to emulate the sensing capabilities of CRISTAL, enabling the detection of sea ice mass redistribution during winter storms. We will integrate multifrequency altimetry and synthetic aperture radar (SAR) to detect snapshots of the 3D sea ice mass balance before and after dynamic events. From a unique database of such events, we will use machine learning to optimize the sea ice ridging scheme in state-of-the-art sea ice models, tailoring them to a new Arctic dominated by weaker seasonal sea ice. We will discover how local episodes of mass imbalance shape the entire Arctic sea ice cover through basin-scale modelling experiments, using one of the most advanced sea ice dynamics models available, neXtSIM. DynAMIC will build Norwegian capacity in sea ice observation-model integration and generate 3D sea ice observations from space required by the Arctic marine shipping, tourism, and prospecting sectors, as well as by interdisciplinary climate researchers.