Unprecedented decline of Arctic sea ice outflow in 2018

The Fram Strait, the primary gateway between the Arctic and North Atlantic Oceans, accounts for approximately 90% of Arctic sea ice export. This outflow significantly influences the freshwater cycle connecting the Arctic and subarctic regions, impacting Nordic Seas stratification, European weather patterns, North Atlantic deep water formation, and global climate. The Arctic Ocean receives 11% of global river discharge into only 3% of the global ocean area, making sea ice export, along with low-salinity seawater, crucial for returning excess freshwater to lower latitudes. This freshwater export, primarily occurring in the Atlantic sector of the Arctic, sets up surface stratification in the Nordic Seas and further downstream. The freshwater delivered to these seas, in both liquid and sea ice forms, preconditions the marine ecosystem, affects European weather and climate, and modulates the rate of deep water formation in the North Atlantic. Historically, sea ice export through Fram Strait mirrored the liquid freshwater transport in the East Greenland Current (EGC). However, recent decades have witnessed a decline in this export, mainly due to a decrease in mean ice thickness in Fram Strait, estimated at ~15% per decade. Climate models predict a continued decline in the coming decades, with significant implications for the Arctic and subarctic hydrological cycles. Sea ice export and liquid freshwater export through the western Fram Strait, approximately from the Prime Meridian to East Greenland's continental shelf, exhibit strong seasonal variations. Within and north of the strait, the EGC interacts with returning warm, saline Atlantic Water (AW), which circulates westward and may subduct under fresher Polar Water, flowing southward again with the EGC. Sea ice in Fram Strait originates from the Siberian shelves (Laptev Sea, East Siberian Sea, and Chukchi Sea) and the southern perimeter of the central Arctic's perennial sea ice cover. The Transpolar Drift stream, driven by anticyclonic winds in the Canada Basin, transports ice floes to Fram Strait, where they continue southward with the EGC, partially driven by northerly winds. This study focuses on the unprecedented decline in sea ice export through Fram Strait in 2018, analyzed using ice thickness measurements from Upward Looking Sonar (ULS), satellite remote sensing, backward trajectories of ice floes, and atmospheric reanalysis data.

Previous research has extensively documented the importance of sea ice export through Fram Strait in the Arctic freshwater cycle and its cascading effects on regional and global climate. Studies such as Peterson et al. (2006) and Carmack et al. (2015) highlighted the trajectory shifts in the Arctic and subarctic freshwater cycle and the role of freshwater in the Arctic marine system. The impact on European climate has been linked to North Atlantic variability (Moffa-Sànchez & Hall, 2017; Woollings et al., 2012). Serreze et al. (2006) and Haine et al. (2015) provided insights into the large-scale freshwater cycle of the Arctic and the mechanisms of freshwater export. The reduction in sea ice export in recent decades has been linked to reduced mean ice thickness (Spreen et al., 2020), and climate simulations predict its continued decline (Holland et al., 2007; Jahn & Laiho, 2020). The role of the East Greenland Current (EGC) in freshwater transport (Aagaard & Coachman, 1968; de Steur et al., 2018; Foukal et al., 2020) and the interaction between the EGC and Atlantic Water (AW) (Hattermann et al., 2016; von Appen et al., 2016) have also been investigated. The origin and transport of sea ice from the Siberian shelves to Fram Strait (Hansen et al., 2013; Krumpen et al., 2019) and the influence of atmospheric circulation (Vinje, 2001; Tsukernik et al., 2010) have been studied. Previous research provides a foundation for understanding the context of the 2018 event and its significance.

This study utilized multiple data sources and analytical techniques to investigate the unprecedented decline in sea ice export through Fram Strait in 2018. Ice thickness data were obtained from Upward Looking Sonars (ULS) and Ice Profiling Sonars (IPS) moored in the East Greenland Current (EGC) at 78.8°N. The moorings provided near-continuous measurements of ice thickness throughout all seasons. Passive microwave satellite observations provided sea ice concentration and drift speed data, covering the entire Arctic and subarctic regions since 1979. However, satellite-based ice thickness estimates were limited to the freezing season due to limitations in measuring ice surface properties during the melting season. To calculate sea ice volume export, the researchers combined ice thickness, sea ice concentration, and ice drift speed. The Fram Strait Arctic Outflow Observatory (AOO), maintained by the Norwegian Polar Institute, provided long-term monitoring of ice thickness and ocean properties in the EGC. In 2018, the ULS array showed an unprecedented decline in ice draft (from which ice thickness is calculated), starting in August 2017 and lasting until August 2018. The decline was accompanied by reduced sea ice concentration, particularly in the eastern EGC. The reduced ice thickness and concentration resulted in a significant reduction in sea ice volume. To investigate the cause of the decline, the researchers constructed backward trajectories of sea ice floes in Fram Strait, analyzing temporal ice thickness evolution along the trajectories, and concurrent thermodynamic forcing on the ice floes. They also used atmospheric reanalysis data from the European Center for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) to study the atmospheric conditions during the period. The sea level pressure (SLP) gradient across the strait and its impact on prevailing wind direction were analyzed. Additionally, salinity data from shipboard CTD measurements were used to examine the effect of the ice thickness reduction on the liquid freshwater content in Fram Strait. The uncertainty estimates for sea ice volume transport were calculated considering errors in ice thickness, concentration, and drift speed, accounting for both systematic and random errors. Backward trajectories were calculated using daily sea ice motion vectors, terminating when ice drift data was unavailable or ice concentration was below 15%. Position errors were assessed by comparing calculated backward trajectories with buoy tracks from the International Arctic Buoy Program (IABP).

The study's key findings reveal an unprecedented decline in sea ice volume transport through Fram Strait in 2018. The annual volume transport of 590 km³ yr⁻¹ represented a 60% reduction compared to the 2000–2017 average and a 75% reduction relative to the 1990s. This minimum was significantly lower than the previous minimum recorded in 2013 (following the 2012 record minimum Arctic summer sea ice extent). Backward trajectory analysis indicated that the ice floes in 2018 followed similar pathways to previous years. However, the ice started to thin rapidly three months before arriving at Fram Strait, unlike in previous years where thickening continued into mid-winter. This rapid thinning coincided with anomalously warm atmospheric conditions during the preceding winter (October 2017–February 2018), with monthly mean 2 m air temperatures nearly 10 degrees higher than the average of the preceding years. In 2018, ice floes remained closer to Fram Strait, spending two to three times longer in an area characterized by a strong heat supply from underlying AW. This longer residence time contributed significantly to the accelerated thickness decline. The slower ice motion and longer residence time were attributed to a repeated reversal of the SLP gradient across the strait and associated changes in wind direction. An east-west dipole SLP anomaly repeatedly appeared, particularly in February 2018, leading to southerly winds that slowed and halted southward ice motion. These southerly winds also carried warm air masses to the area north of Fram Strait, further hindering ice growth in winter. Analysis of a salinity section in Fram Strait in May 2018 showed fresher conditions in the surface layer compared to previous years. However, the observed salinity anomaly was less than one-third of the potential salinity change due to meltwater input, suggesting that much of the meltwater remained upstream, was mixed down, or was not exported simultaneously.

The findings demonstrate that the unprecedented decline in sea ice volume export through Fram Strait in 2018 was primarily driven by a combination of factors. Anomalous atmospheric circulation patterns, characterized by a repeated east-west dipole SLP anomaly, resulted in southerly winds that slowed southward ice drift. This allowed for an extended residence time of sea ice floes north of Fram Strait, exposing them to intense oceanic heat fluxes from the AW and contributing to accelerated melting. The reduced sea ice export has implications for several processes, including: changes in atmosphere-sea ice-ocean interactions, ocean density contrasts, and climate in both the Arctic Ocean and downstream North Atlantic. While the meltwater from the increased melting upstream of Fram Strait might propagate further south with the EGC, potentially influencing dense water formation in the North Atlantic, a significant portion of it seems to remain upstream, be mixed down, or not be exported simultaneously. A considerable portion of sea ice passing through Fram Strait usually drifts into the Nordic Seas, while liquid freshwater primarily follows the Greenland shelf break. The reduced sea ice volume in the EGC might lead to further retreat of the sea ice edge in the Greenland and Iceland Seas, causing stronger air-sea interactions and modifying water masses.

This study reveals an unprecedented decline in Arctic sea ice export through Fram Strait in 2018, primarily attributed to anomalous atmospheric circulation leading to southerly winds and extended sea ice exposure to oceanic heat. The findings underscore the importance of regional atmospheric variability in addition to basin-wide ice thinning. Future research should investigate the long-term impacts of the 2018 event, the likelihood of recurrences given climate change, and the feedback mechanisms between atmospheric patterns and sea ice dynamics.

The study's reliance on ULS/IPS data from a specific location in the Fram Strait might limit the generalizability of the findings to the entire strait. The backward trajectory analysis has inherent uncertainties associated with the accuracy of the ice drift data. Further analysis incorporating data from broader spatial and temporal coverage could improve the accuracy and robustness of the findings.