Volume 123, Issue 3 p. 1586-1594
Introduction to a Special Section
Free Access

Atmosphere-Ice-Ocean-Ecosystem Processes in a Thinner Arctic Sea Ice Regime: The Norwegian Young Sea ICE (N-ICE2015) Expedition

Mats A. Granskog

Corresponding Author

Mats A. Granskog

Norwegian Polar Institute, Fram Centre, Tromsø, Norway

Correspondence to: M. A. Granskog, [email protected]Search for more papers by this author
Ilker Fer

Ilker Fer

Geophysical Institute, University of Bergen, Bergen, Norway

Bjerknes Centre for Climate Research, Bergen, Norway

Search for more papers by this author
Annette Rinke

Annette Rinke

Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Potsdam, Germany

Search for more papers by this author
Harald Steen

Harald Steen

Norwegian Polar Institute, Fram Centre, Tromsø, Norway

Search for more papers by this author
First published: 28 February 2018
Citations: 40


Arctic sea ice has been in rapid decline the last decade and the Norwegian young sea ICE (N-ICE2015) expedition sought to investigate key processes in a thin Arctic sea ice regime, with emphasis on atmosphere-snow-ice-ocean dynamics and sea ice associated ecosystem. The main findings from a half-year long campaign are collected into this special section spanning the Journal of Geophysical Research: Atmospheres, Journal of Geophysical Research: Oceans, and Journal of Geophysical Research: Biogeosciences and provide a basis for a better understanding of processes in a thin sea ice regime in the high Arctic. All data from the campaign are made freely available to the research community.

Key Points

  • An interdisciplinary expedition in the Atlantic sector of the high-Arctic studied the dynamics of a thin (<1.5 m) sea ice regime
  • Observations in a rarely sampled region, especially in winter, highlight the conditions in the Atlantic sector of the Arctic basin
  • Especially short-lived passing storms, with high winds and air temperatures and precipitation, leave a lasting impact on the sea-ice system

1 Background and Motivation

Arctic sea ice has changed rapidly in the last decade. The sea ice extent has declined (e.g., Meier et al., 2014), coincidently ice thickness has decreased, nearly halved from about 3 m a few decades ago (e.g., Hansen et al., 2013; Lindsay & Schweiger, 2015). The sea ice has transformed from a thicker pack with predominantly older multiyear ice (MYI) into an ice pack of predominantly younger and thinner first-year sea ice (FYI) (e.g., Maslanik et al., 2011).

The Surface Energy Balance of the Arctic Ocean (SHEBA) experiment (Perovich et al., 1999) has shaped our understanding of the physics of the Arctic sea ice tremendously. The 1 year long experiment took place in 1997–1998. Most recent spur of activity during the International Polar Year (IPY) in 2007–2008, included the transpolar drift of schooner Tara (Gascard et al., 2008), while the Circumpolar flaw lead study (CFL-IPY) (Barber et al., 2010) focused on processes at the margins of the ice pack north of Canada. All these experiments took place when thicker perennial ice was more prevalent in the Arctic basin. Thus much of our knowledge of the Arctic sea-ice system stems from observations that were carried out during an era prior to the rapid change experienced in the last decade.

This brings forward the question how atmosphere-ice-ocean interactions function in the thinner ice regime compared to the thicker ice pack only two decades ago. How do the changes in the sea ice pack and interaction with the atmosphere and ocean affect the functioning of the marine ecosystem (e.g., Meier et al., 2014)? In fact quite little is known. For example, the ice pack has become more mobile (Spreen et al., 2011), and thinning may have increased the amount of sunlight reaching the ocean (e.g., Nicolaus et al., 2012). Snow depth on first-year sea ice in the western Arctic has decreased (Webster et al., 2014), although precipitation in the Arctic could increase partly due to sea-ice retreat (e.g., Bintanja & Selten, 2014). What other shifts or changes can we expect in a thinner ice pack and do these findings apply in the sparsely sampled region north of Svalbard? Overall, our ability to understand and foresee future is hampered by scarcity of data.

The Norwegian young sea ICE (N-ICE2015) experiment was designed to study the functioning of the thinner drifting Arctic sea ice pack and its interactions with the atmosphere and ocean, and effects on the marine ecosystem (Granskog et al., 2016). The aim was to provide new process understanding, to provide new data sets and make them freely available to the research community to enhance synergy and scientific impact, and to evaluate numerical models and remote sensing data and thus to provide pathways to improve climate models. An interdisciplinary systems approach was fundamental to the science of N-ICE2015.

The overarching objective of this paper is to introduce the study, note on the main findings from the different components of the atmosphere-snow-ice-ocean system and refer the reader to the specific articles contained in this special section.

2 Field Campaign

At the core of N-ICE2015 was a five and a half month drift study in the Arctic sea ice pack north of Svalbard. This region is characterized by two major features, warm Atlantic water inflow to the Arctic Ocean along the continental slope and the transport of sea ice with the Transpolar Drift System from the eastern Arctic. Sea ice in the Transpolar Drift System has been thinning in the last decade. For example, in Fram Strait the long-term observations (1990–2011) indicate a significant thinning of modal (level) ice thickness by 32%, after 2007 compared to the 1990s (Hansen et al., 2013). Thus, the study location was suitable to examine the functioning of a younger and thinner ice pack, than for example during SHEBA.

The observations were designed to capture key processes related to the interaction of the sea-ice pack with atmosphere and ocean, to quantify the forcing from atmosphere and ocean, and to observe the snow and ice thermodynamics and dynamics in response to the forcing. The lower trophic levels and the biogeochemistry of the upper ocean and sea ice were followed throughout the campaign to understand linkages with the dynamics of the physical environment and ecosystem functioning (Granskog et al., 2016). Observations were collected from the midst of the polar night (winter) in mid-January through the advancement of spring, and to early summer in June, as this seasonality is one of the characteristics of the Arctic system.

In addition to ship-based observations and the main ice camp, autonomous platforms were used, to provide improved spatial and temporal coverage in the study region. Airborne campaigns, such as NASA's Operation Icebridge and EU project ICE-ARC, used the opportunity to get valuable ground-truth data from drift ice. Similarly the ground-truth data provided valuable information for validation of satellite remote sensing of sea ice and aerosols.

At the outset, the experiment was planned as two drifts of 2–3 months duration each with the research vessel RV Lance as the main base, starting at approximately 83°N. But the conditions with a combination of rapid ice drift, severe storms, swell, and the breakup of ice floes forced us to relocate the ice camps several times, and sooner than expected. Thus the experiment consisted of four separate drifts (Figure 1), where the three first (Floes 1, 2, and 3) began at about 83°N in the deep parts of the Nansen Basin, while the last drift (Floe 4) took place closer to the ice edge due to time constraints (Figure 1).

Details are in the caption following the image

Map showing the four different Floe drifts during the N-ICE2015 campaign.

Floes 1 and 2 covered the winter, Floe 3 the spring, and Floe 4 the early summer (Cohen et al., 2017), with snow melt onset in the area being in early June. The typical ice thickness in the study region was <1.5 m with a relatively thick snow cover (0.3–0.5 m) (King et al., 2016; Rösel et al. 2016a, 2016b, 2018). Based on back tracking of the sea ice (Itkin et al., 2017), the oldest ice in the ice pack was inferred to be second-year ice (SYI) originated from the Laptev Sea in fall 2013. Thus the campaign was undertaken in an ice pack significantly thinner than SHEBA or the Tara drift.

3 Overview of Conditions During Experiment and Main Findings

3.1 Atmospheric Conditions

In winter, the atmospheric conditions were characterized by several intense storms that entered the Arctic via the Greenland Sea and passed the N-ICE2015 ice camp, in response to the prevailing large-scale atmospheric circulation pattern with an anomalous deep vortex shifted over eastern North America and meandering jet stream over the Arctic North Atlantic (Cohen et al., 2017). The spring/summer storms were less intense and traveled north through the Barents and Kara Seas (Cohen et al., 2017). The passing storms made a significant impact, for example, on the surface energy budget (Walden et al., 2017), the entire tropospheric column and boundary layer characteristics (Kayser et al., 2017), sea ice deformation and drift (Itkin et al., 2017), and upper ocean mixing and thus turbulent ocean heat fluxes (Fer et al., 2017; Meyer et al., 2017a, 2017b; Peterson et al., 2017). Winter observations and direct observations from Arctic storms are rare, and the N-ICE2015 campaign provided the first direct integrated observations of the effect of such storms in the Arctic Basin.

The observed intense storms are related to midwinter warming events (temperature increase of over 20 K in less than 24 h) and are a normal part of the Arctic climate, but they are becoming more frequent and lasting longer than they did three decades ago (Graham et al., 2017a; Rinke et al., 2017). Storm events are associated with transitions from the radiatively clear (i.e., cold, strongly stable stratified) to the opaquely cloudy (i.e., warmer) state (Graham et al., 2017b; Stramler et al., 2011). A comparison of the atmospheric conditions during N-ICE2015 and SHEBA revealed that they were remarkably similar (Graham et al., 2017b; Kayser et al., 2017; Walden et al., 2017), and suggests these two states are an Arctic-wide phenomenon (Graham et al., 2017b). However, most climate models have difficulties to represent them adequately (Graham et al., 2017b; Pithan et al., 2014). Using ERA-Interim re-analysis data Graham et al. (2017b) showed that there is a positive trend for the number of the (warmer) opaquely cloudy days in the N-ICE2015 study area, consistent with an increased number of moisture intrusions in the Atlantic sector (Woods & Caballero, 2016). However, the changes in the occurrence of both states are not uniform across the Arctic Basin.

Using lidar observations from ice-tethered buoys during N-ICE2015 and other deployments north of Svalbard, Di Biagio et al. (2018) studied aerosol properties and made comparisons to the CALIOP (Cloud and Aerosol Lidar with Orthogonal Polarization) data. They conclude that in winter dusty-type aerosol (dusty marine, desert dust, polluted dust) seems to be overrepresented in the CALIOP data, while buoy observations indicate these could be more likely associated to diamond dust (tiny ice crystals). A trajectory analysis indicates that the Arctic is a pathway for aerosol between Russia/Europe and North America.

Sparse Arctic observations limit the initialization of weather prediction models and thus the accuracy of the prediction not only over the Arctic but also over midlatitudes. Sato et al. (2017) showed that the assimilation of the additional radiosonde observations during N-ICE2015 (Kayser et al., 2017) helped to improve the forecasts of weather events in North America as well as in East Asia in winter 2015.

3.2 Hydrography and Ocean Dynamics

The observations of hydrography, ocean currents and mixing in the water column, together with under-ice boundary layer measurements provided unique data sets and new insights in a region with scarce prior observations, especially in winter (Koenig et al., 2016, 2017; Meyer et al., 2017a, 2017b; Peterson et al., 2017). Episodic events during storms were found to be important for heat, salt, and nutrient fluxes (Fer et al., 2017; Koenig et al., 2016; Meyer et al., 2017a; Peterson et al., 2017; Randelhoff et al., 2016). Ocean heat fluxes were affected by a complex interplay of wind forcing, topography, and proximity to the warm Atlantic Water (AW) pathways (Meyer et al., 2017a), whereas the nutrient fluxes were strongly affected by the evolution of upper ocean stratification and meltwater layers (Randelhoff et al., 2016). Largest ocean heat fluxes exceeded 300 W m−2 when drifting over AW during storms, resulting in rapid ice melt (Meyer et al., 2017a; Peterson et al., 2017; Provost et al., 2017). A combination of observations and modeling (Koenig et al., 2017; Meyer et al., 2017b) added to our understanding of the AW pathways in the Yermak Plateau (YP) area.

The winter hydrography in the Nansen Basin and over the flanks of the YP was characterized by a deep (up to 100 m) mixed layer and a strong pycnocline (Meyer et al., 2017b). In late spring, closer to the ice edge over the YP, the mixed layer shoaled to less than 20 m. Profiling buoys supplemented the sampling in winter (Koenig et al., 2016; Provost et al., 2017). The AW inflow north of Svalbard was found to be steered by topography, partly along the Svalbard coast and partly around the YP. Using profiling buoys, Koenig et al. (2016) suggest a contribution of warm AW inflow through the Yermak pass, a 700 m deep passage over the YP. Numerical operational model outputs support these interpretations of warm water pathways (Koenig et al., 2017).

Oceanic turbulent heat flux measurements were made in the under-ice boundary layer using eddy-covariance instrument clusters (Peterson et al., 2017), and in the upper 200–300 m using loosely tethered microstructure profilers (Meyer et al., 2017a). While the under ice measurements quantify the ocean-to-ice heat flux accurately (Peterson et al., 2017), the microstructure observations were used to characterize the vertical mixing across the pycnocline (Meyer et al., 2017a). In the Arctic winter, the only source of heat to sea ice is the ocean. Oceanic heat fluxes measured 1 m below the sea ice in the Nansen Basin were 1 W m−2 in winter, and increased by a factor of two during wind events (Peterson et al., 2017). Fluxes inferred from the water column were slightly larger, suggesting heat flux convergence toward the sea ice, but the increase in heat fluxes from calm to stormy conditions were similar: winter heat flux across the pycnocline in the Nansen Basin averaged to 3 W m−2 during calm conditions and increased to 5 W m−2 with storms. In spring, the combination of radiation absorbed in the upper ocean, wind forcing, shallow AW layer and proximity to open waters lead to rapid melting and large ocean-to-ice heat fluxes exceeding several 100 W m−2 (Peterson et al., 2017; Provost et al., 2017). Comparably large heat fluxes were observed across the pycnocline (Meyer et al., 2017a). As the ice camp drifted over the continental slope, steep topography enhanced dissipation rates by a factor of four along the eastern slopes of the YP, and episodically increased the turbulent heat flux deeper in the water column. Fer et al. (2017) used winter data to describe the evolution of the upper ocean hydrography by one-dimensional processes, forced by idealized, time-dependent vertical diffusivity profiles inferred from microstructure measurements. The increase in the mixed-layer salinity was dominantly from entrainment of saline AW layer from below, particularly during episodic strong wind forcing, whereas salinification from brine release during freezing was only 10%.

Taskjelle et al. (2017) combined optical measurements with radiative transfer modeling during an under-ice phytoplankton bloom in May to show that the mean total absorption in the upper 20 m of the water column increased by a factor of 4, resulting in 35% more solar energy deposited in the upper 10 m. Randelhoff et al. (2016) combined the N-ICE2015 data set with three earlier surveys (winter, spring, and summer), covering different regimes from open waters in Fram Strait to the pack ice north of Svalbard. All data sets included colocated continuous vertical profiles of nitrate concentrations and vertical diffusivity from ocean microstructure profiles, and were used to infer vertical fluxes of nutrient, their seasonal evolution, and relation to upper ocean stratification. Such observations are crucial to quantify the contribution of the oceanic nutrient reservoir to “new” production. New production is supported by nutrient inputs from outside the euphotic zone, as opposed to “regenerated” production, which is supported by recycling of nutrients in the euphotic zone. The upper ocean nitrate drawdown was found to be linked to the stratification formed from sea ice melt. Meltwater layers significantly reduce the turbulent mixing in upper ocean (Randelhoff et al., 2017), with important consequences for vertical fluxes. During summer, upward nitrate fluxes across the seasonal nitracline increased twofold in open water, relative to observations under sea ice cover. The suppressed levels under ice were attributed to the development of stratification from continued input of meltwater. Randelhoff and Guthrie (2016) combined the N-ICE2015 observations with existing data sets (of nutrients and diffusivity) from other parts of the central Arctic Ocean to quantify the circum-Arctic export production. The average overall new production exported across the nitracline was estimated to be in the range 1.5–3 g C m−2 yr−1.

3.3 Snow and Sea Ice

3.3.1 Mass Balance and Snow Cover

The observations of ice thickness during the N-ICE2015 campaign confirm that the ice was thinner than from historical records in the region. Total ice and snow thickness was about 1.6–1.7 m, less than the earlier observations of 1.8–2.7 m in the same region at same time of year (Rösel et al., 2018). Given the snow depth was order of 0.3–0.5 m, the ice thickness was typically only 1.1–1.4 m, this is much thinner than the typical modal ice thickness that was about 2.0 m at end of winter during SHEBA (Eicken et al., 2001; D. K. Perovich, personal communication, 2018).

Thick snow (0.3–0.5 m) was found on both first-year and older ice (Gallet et al., 2017; Merkouriadi et al., 2017b; Rösel et al., 2016a, 2018). Already in January, the snow pack was about 0.5 m thick on SYI (Merkouriadi et al., 2017a; Rösel et al., 2016a, 2018), much thicker than one could expect from climatology (Warren et al., 1999) and in contrast to the reports of thinning snow on sea ice in the western Arctic (Webster et al., 2014). In recent years, observations have pointed toward rather thick snow in the area north of Svalbard (Haapala et al., 2013), although scarce data in the region hint toward increasing snow depth on sea ice north of Svalbard (Rösel et al., 2018).

Thick snow limited thermodynamic bottom ice growth in winter and spring (Rösel et al., 2018). Limited ice growth was corroborated by the observations in the ocean that indicated that brine fluxes from sea ice were small (Fer et al., 2017). Two mechanisms contributed to flooding of the sea ice and snow-ice formation: ice breakup during storms and basal ice melt when the sea ice drifted into the main pathways of warmer Atlantic waters north of Svalbard (Provost et al., 2017). Sea ice cores indicated that snow contributed significantly to the mass balance of SYI (almost 10% by mass and 20–30% of thickness) (Granskog et al., 2017). These are the first reports of widespread flooding (Rösel et al., 2018) and snow-ice formation for pack ice in the central Arctic, a potential consequence of the thinning of sea ice in the Transpolar Drift (Haas et al., 2008; Hansen et al., 2013) and the regionally higher snow accumulation in the area north of Svalbard and the Barents Sea (Merkouriadi et al., 2017a, 2017b). For SYI in the western Transpolar Drift growth in the preceding winter was by snow-ice formation (Merkouriadi et al., 2017a).

The studies of snow on sea ice were the first comprehensive ones in this sector of the Arctic Ocean (Gallet et al., 2017; Merkouriadi et al., 2017b). In addition to a heavy snow load, the variable weather conditions, e.g., passing storms which affect the thermal gradients in the snow pack, has likely an imprint on the snow pack composition, making the snow pack distinctively different than observed during SHEBA (Merkouriadi et al., 2017b). It merits a further investigations how the amount and properties of the snow in this region affects sea-ice growth.

3.3.2 Sea Ice Dynamics

Initial analysis show the ice drift speeds in the study area have increased in the period from 2002 to 2016 (Itkin et al., 2017). One of the larger efforts during N-ICE2015 was to use autonomous buoys to understand the dynamics of the thinner ice pack (Itkin et al., 2017). These buoys also provided a more continuous record of snow and ice evolution in the study area (Provost et al., 2017; Rösel et al., 2018). In total over 40 different buoys of different types were deployed during N-ICE2015 distributed in two arrays, one in winter and one in spring (Itkin et al., 2015). Results indicate that there is an increase in deformation associated with the younger and thinner sea ice, and signs of a destructive role of winter storms that preconditions the ice pack (Itkin et al., 2017).

In addition to buoys, the ship's radar was used to observe sea ice dynamics on a smaller scale (within 15 km of the ship), using virtual buoys (Karvonen, 2016). The results from this work showed that breakup of the ice often occurred along earlier faults, which had been weakened during earlier dynamic events often associated with storms (Oikkonen et al., 2017).

Using high-resolution airborne data on sea-ice freeboard before and after a passing storm, Itkin et al. (2018) showed in detail how such a storm redistributes ice mass, and creates opportunity for new ice growth in leads, in absence of thermodynamic thickening of ice covered by thick snow.

3.3.3 Sea Ice Remote Sensing

The N-ICE2015 experiment also provided a valuable platform for validation of remote sensing products, from both satellite and airborne observations. Unique near-coincident synthetic aperture radar (SAR) data at multiple frequencies and simultaneous ground truth could be gathered, that provided new insights into use of imagery for detection of open water and thin ice and for sea ice classification (Espeseth et al., 2016; Johansson et al., 2017, 2018; Ressel et al., 2016; Rösel et al., 2017).

King et al. (2018) showed that radar reflections from both CryoSat-2 and an airborne radar instrument were closer to snow freeboard than ice freeboard, resulting in a systematic overestimation of sea-ice thickness in the N-ICE2015 region in spring 2015 by a factor of two in all operational CryoSat-2 products of date, despite of low temperatures (below −15°C). This is likely a result of the snow properties at the time in the region and highlights the need to be careful when interpreting CryoSat-2 data in this region.

3.4 Biogeochemistry and Ecosystem Processes

The thick snow cover also had an impact on the biomass build up in sea ice. It was rather low in older SYI ice with thick (0.5 m) snow (Olsen et al., 2017), while in thinner young ice formed in refrozen leads, with much less snow (<0.1 m), ice algal standing stocks quickly approached similar levels as in the older ice (Kauko et al., 2017). On the other hand, the ice algae in the refrozen leads were exposed to much higher light levels and had to produce large amount of sunscreens, so called mycosphorine amino acids (MAAs), at highest levels observed in sea ice to date (Kauko et al., 2017). The older ice may be a safe haven for ice algae that seed algae growth in younger ice types in spring (Olsen et al., 2017), the loss of older ice could have direct consequences for the sea-ice ecosystem. The ice algae also contributed significantly to the absorption of sunlight in the sea ice cover (Kowalczuk et al., 2017).

Assmy et al. (2017) integrated observations of the atmosphere, sea ice, and ocean during N-ICE2015 to understand the dynamics of an early spring under-ice phytoplankton bloom, dominated by the algae Phaeocystis pouchetii, which thrived during the drifts of Floes 3 and 4. This bloom was observed much earlier in the season and much further north than previous under-ice blooms below ponded ice in late summer that usually were dominated by diatoms (e.g., Arrigo et al., 2012). Enough leads developed, with open water or thin snow-free ice, that enough light passed to the ocean below to sustain phytoplankton growth, despite virtually no light passed through the FYI and SYI ice with thick snow (Taskjelle et al., 2017). The under-ice bloom increased absorption and scattering of sunlight (Pavlov et al., 2017) affected the vertical deposition of solar heat thus potentially contributing to ice melt (Taskjelle et al., 2017).

The collected time series on sea-ice physics and biogeochemistry further provided an opportunity to evaluate a state of the art sea-ice numerical model (Duarte et al., 2017), showing that physics can be well reproduced given the forcing is realistic, and also giving insights into ice algae dynamics in different types of sea ice. A unique time series quantified the factors that affect surface water CO2, showing the importance of transient storm events and ice dynamics for ocean-atmosphere CO2 exchange in ice-covered waters (Fransson et al., 2017). While Nomura et al. (2018) showed that the young new sea ice growing in winter and spring is a source CO2 to the atmosphere, while in older ice the CO2 flux is strongly modulated by the snow cover properties.

4 Summary

The N-ICE2015 campaign provided novel interdisciplinary data from the high Arctic and new insights into the functioning of an Arctic sea-ice pack that was relatively thin. This special section contains 29 articles that summarize the main findings from this campaign.

A recurring theme in many of the findings is the impact of the frequent storms that passed through the study area during the experiment. Unique direct observations of the immediate and longer term effects of storms on atmosphere-ice-ocean interaction are described. Storms entering the Arctic Ocean from the Atlantic are characteristics for this region, setting this region apart from other parts of the Arctic. The thin ice pack is more vulnerable to external forcing, as exemplified by its dynamical response to winds, and this may have lasting effects on the functioning of the ice pack.

The regionality in the conditions across the Arctic need to be kept in mind when undertaking studies in the Arctic, as not all findings can be extrapolated to be representative of the whole Arctic even if we wish that is the case. Observations from all different parts of the Arctic are needed, as no one region is representative of the whole Arctic.

To unravel complex interactions in the Arctic sea-ice system long-term campaigns, which cover the crucial seasonality and associated feedbacks, are needed, thus the forthcoming Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) in 2019–2020, in the central Arctic, comes in nick of time, but one has to be cautious when extrapolating findings from such experiments to the whole Arctic.


The N-ICE2015 campaign was supported by the Centre of Ice, Climate and Ecosystems (ICE) at the Norwegian Polar Institute through the N-ICE project. We acknowledge support from The Ministry of Climate and Environment and the Ministry of Foreign Affairs of Norway. Support was also provided through the ICE-ARC (Ice, Climate, Economics-Arctic Research on Change) programme from the European Union 7th Framework Programme, grant 603887, and the Centre for Climate Dynamics at the Bjerknes Centre through the BASIC project, and several Research Council of Norway projects, and the European Space Agency. With the contributions from the Alfred Wegener Institute (AWI) and Korean Polar Research Institute (KOPRI) the radiosonde program was made possible. We like to especially thank the captains and crews of RV Lance for the support during the challenging conditions during the field campaign. Logistics, safety training and general support from the Operations and Logistics department at the Norwegian Polar Institute are greatly acknowledged. Special thanks also to the captains and crews of the Norwegian Coast Guard vessel KV Svalbard who supported the transits from the ice edge to 83°N. We thank the Governor of Svalbard (Sysselmannen) for support with helicopter operations and the helicopter crews. We like to thank all the Scientific Editors and numerous reviewers who have made a great effort to ensure the quality of the work published in this special issue. Last but not least, we like to thank all the scientists involved. They had to work in harsh conditions in the field to collect the raw data, and have done an impressive job in a relatively short time to publish the results in this special section and beyond. The data from this campaign are made available through the Norwegian Polar Data Centre (data.npolar.no) and other data repositories. Individual data sets are cited accordingly in the individual papers of this special section.