Arctic sea ice has displayed significant thinning as well as an increase in drift speed in recent years. Taken together this suggests an associated rise in sea ice deformation rate. A winter and spring expedition to the sea ice covered region north of Svalbard–the Norwegian young sea ICE2015 expedition (N‐ICE2015)—gave an opportunity to deploy extensive buoy arrays and to monitor the deformation of the first‐year and second‐year ice now common in the majority of the Arctic Basin. During the 5 month long expedition, the ice cover underwent several strong deformation events, including a powerful storm in early February that damaged the ice cover irreversibly. The values of total deformation measured during N‐ICE2015 exceed previously measured values in the Arctic Basin at similar scales: At 100 km scale, N‐ICE2015 values averaged above 0.1 d⁻¹, compared to rates of 0.08 d⁻¹ or less for previous buoy arrays. The exponent of the power law between the deformation length scale and total deformation developed over the season from 0.37 to 0.54 with an abrupt increase immediately after the early February storm, indicating a weakened ice cover with more free drift of the sea ice floes. Our results point to a general increase in deformation associated with the younger and thinner Arctic sea ice and to a potentially destructive role of winter storms.

We explore new and existing convolutional neural network (CNN) architectures for sea ice classification using Sentinel-1 (S1) synthetic aperture radar (SAR) data by investigating two key challenges: binary sea ice versus open-water classification, and a multi-class sea ice type classification. The analysis of sea ice in SAR images is challenging because of the thermal noise effects and ambiguities in the radar backscatter for certain conditions that include the reflection of complex information from sea ice surfaces. We use manually annotated SAR images containing various sea ice types to construct a dataset for our Deep Learning (DL) analysis. To avoid contamination between classes we use a combination of near-simultaneous SAR images from S1 and fine resolution cloud-free optical data from Sentinel-2 (S2). For the classification, we use data augmentation to adjust for the imbalance of sea ice type classes in the training data. The SAR images are divided into small patches which are processed one at a time. We demonstrate that the combination of data augmentation and training of a proposed modified Visual Geometric Group 16-layer (VGG-16) network, trained from scratch, significantly improves the classification performance, compared to the original VGG-16 model and an ad hoc CNN model. The experimental results show both qualitatively and quantitatively that our models produce accurate classification results.

Renewed political and commercial interest in the resources of the Arctic, the reduction in the extent and thickness of sea ice, and the recent failings that led to the Deepwater Horizon oil spill, have prompted industry and its regulatory agencies, governments, local communities and NGOs to look at all aspects of Arctic oil spill countermeasures with fresh eyes. This paper provides an overview of present oil spill response capabilities and technologies for ice-covered waters, as well as under potential future conditions driven by a changing climate. Though not an exhaustive review, we provide the key research results for oil spill response from knowledge accumulated over many decades, including significant review papers that have been prepared as well as results from recent laboratory tests, field programmes and modelling work. The three main areas covered by the review are as follows: oil weathering and modelling; oil detection and monitoring; and oil spill response techniques.

The Norske Øer Ice Barrier (NØIB) is a region of fast ice located off the northeast coast of Greenland. It is one of the most extensive areas of landfast ice on Earth. This paper looks at the NØIB formation during the freeze-up of late 2003 and the break-up in summer 2004. As the fast ice is immobile, it provides an ideal location for checking the consistency of classification schemes for satellite sensors. Active microwave (SAR) backscatter values from Envisat are compared with optical observations from the MODIS, multichannel passive microwave from the SSM/I and with ice-freeboard values from the Envisat RA-2. In August 2004 the underside of the NØIB was mapped by an upward-looking multibeam sonar mounted on the Autosub autonomous underwater vehicle. Statistics from sea-ice draft measurements by the multibeam are compared with near-coincident satellite observations. Evaluating the evolution of the fast ice through multiple satellite sensors with ground truth measurements may allow future development of improved automatic classification algorithms which will be better able to track fast-ice extent. Loss of the fast ice for periods of the year has implications for the coastal environment of Greenland and may contribute to the retreat of the Nioghalvfjerdsfjorden glacier and enhanced coastal erosion.

The most common source of information about sea ice conditions is remote sensing data, especially images obtained from synthetic aperture radar (SAR) and passive microwave radiometers (PMR). Here we introduce an adaptive fusion scheme based on Graph Laplacians that allows us to retrieve the most relevant information from satellite images. In a first test case, we explore the potential of sea ice classification employing SAR and PMR separately and simultaneously, in order to evaluate the complementarity of both sensors and to assess the result of a combined use. Our test case illustrates the flexibility and efficiency of the proposed scheme and indicates an advantage of combining AMSR‐2 89 GHz and Sentinel‐1 data for sea ice mapping.