Winter Intrusions of Atlantic Water in Kongsfjorden: Oceanic Preconditioning and Atmospheric Triggering

Kongsfjorden is an Arctic fjord in Svalbard facing the West Spitsbergen Current (WSC) transporting warm and salty Atlantic Water (AW) through the Fram Strait to the Arctic. In this work, winter AW intrusions in Kongsfjorden occurring in the 2010-2020 decade are assessed by means of oceanographic and atmospheric observations, provided by in-situ instrumentations and reanalysis products. Winter AW intrusions are relatively common events, bringing heat and salt from the open ocean to the fjord interior; they are characterized by water temperatures rising by 1-2 °C in just a few days. Several mechanisms have been proposed to explain winter AW intrusions in West Spitsbergen fjords, tracing back to the occurrence of energetic wind events along the shelf slope. Here we demonstrate that the ocean plays a fundamental role as well in regulating the inflow of AW toward Kongsfjorden in winter.Winter AW intrusions in 2011, 2012, 2016, 2018 and 2020 occurred by means of upwelling from the WSC, triggered by large southerly winds blowing on the West Spitsbergen Shelf (WSS) followed by a circulation reversal with northerly winds. Southerly winds are generated by the setup of a high pressure anomaly over the Barents Sea. In these winters, fjord waters are fresher and less dense than the AW current, resulting in the breakdown of the geostrophic control mechanism at the fjord mouth, allowing AW to enter Kongsfjorden. The low salinity signal is found also on the WSS and hence is related to the particular properties of the Spitsbergen Polar Current (SPC). The freshwater signal is hypothesized to be linked to the sea-ice production and melting in the Storfjorden and Barents Sea regions, as well as the accumulation of glaciers’ runoff. The freshwater transport toward West Spitsbergen is thus the key preconditioning factor allowing winter AW intrusions in Kongsfjorden by upwelling, whilst energetic atmospheric phenomena trigger the intrusions. Winter 2014 AW intrusion shows a different dynamic, i.e., an extensive downwelling of warm waters in the fjord lasting several weeks. Here, long-lasting southerly winds stack surface waters toward the coast. The fjord density is larger than the WSC density, forcing the AW intrusion to occur near the surface, then spreading vertically over the water column due to heat loss to the atmosphere. We hypothesize the combination of sustained Ekman transport and the shallower height of the WSC on the water column to be the key factor explaining the AW intrusion in this winter. After mixing with the initial AW inflow, fjord waters undergo heat loss to the atmosphere and densification. The water column becomes denser than the WSC, restoring the geostrophic control mechanism and blocking further intrusions of AW.

The West Spitsbergen Current (WSC) is the major source of heat and salt for the Arctic Ocean and the areas of deep convection in the Greenland Sea. The WSC current cools dramatically downstream. Hydrographic and velocity data from a 3‐week, midwinter cruise off Spitsbergen are used to investigate the heat budget of the WSC and the mechanisms of cooling. The downstream divergence of mean heat flux in the WSC produces a heat loss of at least 1000±400 Wm−2 averaged over the width of the current. Approximately 350 Wm−2 is lost to the atmosphere and 200 Wm−2 is lost to melting ice over a region somewhat wider than the current. Cooling of the WSC to the atmosphere converts the inflowing Atlantic Water (AW) to Lower Arctic Intermediate Water, which is sufficiently salty to convect. Cooling by ice converts the AW to much fresher Arctic Surface Water, which is too light to convect. The relative importance of these two conversions is primarily controlled by the rate at which the wind advects ice from the Barents Sea over the WSC. The warmest water of the WSC is often observed 100–200 m below the surface. Despite the lack of direct contact with the surface, this warm core cools at about 800 Wm−2 in our observations. This rate is too large to be caused by diapycnal diffusion. We suggest that the energetic eddy field in this area diffuses heat along the steeply sloping isopycnal surfaces that connect the warm core to the surface, renewing the surface layer several times per day. This is consistent with the very shallow surface mixed layers and high level of intrusions observed. We conclude that the surface layer of the WSC is cooled by the atmosphere and by ice from the Barents Sea and that isopycnal diffusion by mesoscale eddies continually renews this surface, thus cooling the interior of the WSC. The relative magnitude of these processes determines whether the inflowing warm, salty AW is converted to light, fresh surface water or salty, cold intermediate water.

The Kongsfjorden-Krossfjorden system, situated on the northwest coast of Spitsbergen, is connected to the continental shelf slope by a trough, Kongsfjordrenna, that crosses the 50 km wide shelf. Kongsfjorden is the southern arm of this fjord system, and has a maximum depth of 400 m. The fjord system has no typical fjord sill, but Kongsfjordrenna seems to function as a sill of around 270 meters. This means that most of the water column in Kongsfjorden is susceptible to exchange with warm and salty Atlantic Water (AW) from the West Spitsbergen Current (WSC) flowing along the shelf slope and with colder and fresher water from the shelf. The water masses found in Kongsfjorden can be viewed on as a mixture between the AW, Winter Cooled Water (WCW) and fresh water either as melt water or river runoff. Winds at the west Spitsbergen coast cause Eknian drift to either pile up or remove surface water from the coast. The result is an altering of the stratification of the water outside the fjord system in the form of downwelling or upwelling. This builds up a horizontal pressure gradient between the coast and the fjord system, forcing water in or out of the fjord area. Conservation of volume demands that exchange takes place in both directions. The exchange with the adjacent shelf is shown to be related to irreversible exchange across the front between the AW along the shelf slope and the ArW on the shelf. Most of the exchanged water crosses the shelf and numerous intrusions to the fjord system take place every year. The marine ecosystem is dependent on such variations, especially on the lower trophic level, such as microbial and zooplankton production. The relative composition of zooplankton depends on water masses and sea ice concentration. Changes in the zooplankton composition will result in altered energy transfer within the pelagic food web with potential consequences for growth and survival of seabirds. This work is based on CTD data from four cruises to Kongsfjorden: last week of April, first week of June, first week of July and third week of September. The volume of AW and freshwater in Kongsfjorden during each of these periods was estimated mainly to investigate the variability in the content of these water types. Estimation of fresh water content due to ice melting and river runoff and the estimation of the amount of fresh water was carried out by subtracting all measured salinities from the maximum salinity measured in the standard shelf slope station The areal network was used to estimate the volume of the mentioned water masses. Variations in the amount of AW and fresh water have been observed between these periods. The comparison of TS profiles at one position in the fjord from all the four cruises, and two profiles from a station in the AW on the shelf slope show some of these characteristics. Evidence that these variations are caused by AW entering the fjord by eddies escaping from the WSC due to instabilities along the front between the WSC and the shelf water has been encounter.

<p>In the last two decades, rising ocean temperatures have significantly contributed to the increased melting and retreat of marine-terminating glaciers along the coast of Greenland. Warming subsurface waters have also been shown to interact with the glaciers in Northeast Greenland, which until recently were considered stable, and caused their rapid retreat. The main source of these waters is the westward recirculation of subducted Atlantic Water (AW) in Fram Strait, which has shown a warming of up to 1° C over the past few decades.</p><p>In this study, the connection between the subsurface warm Atlantic Intermediate Water (AIW) found on the wide continental shelf of Northeast Greenland and in the fjords, and AW within the West Spitsbergen Current (WSC) is investigated using historical hydrographic observations and high-resolution numerical simulations with the Finite-Element Sea-ice Ocean Model (FESOM). We find that AW from the WSC takes between 10 – 14 months to recirculate across Fram Strait and reach the shelf break where it moves southwards. The pronounced inter-annual variability in the WSC is preserved as the water recirculates. However, the variability of temperature and AIW layer thickness on the shelf at seasonal or inter-annual time scales is at best weakly controlled by the AW temperature in the WSC. There is no significant correlation between AIW and the WSC anywhere on the shelf, suggesting advection from the WSC alone does not control AIW signals. The role of wind-driven, episodic upwelling is then investigated as a driver of transport of AIW from Fram Strait onto the shelf (following an approach by Münchow et al., 2020) where it then may follow the deep trough system towards the glaciers.</p>

<p>The Barents Sea is one of the main pathways by which Atlantic Water (AW) enters the Arctic Ocean and is an important region for key water mass transformation and production. As AW enters the shallow (< 400 m) Barents Sea, it propagates as a topographically steered current along a series of shallow troughs and ridges, while being transformed through atmospheric heat fluxes and exchanges with surrounding water masses. To the north, the warm and salty AW is separated from the cold and fresh Polar Water (PW) by a distinct dynamic thermohaline front (the Barents Sea Polar Front), often less than 15 km in width.</p><p>Two cruises were conducted in October 2020 and February 2021 within the Nansen Legacy project, focusing on the AW pathways and ocean mixing processes in the Barents Sea. Here we present data from CTD (Conductivity, Temperature, Depth), ADCP (Acoustic Doppler Current Profiler) and microstructure sensors obtained during seven ship transects and two repeated stations across and on top of a 200 m deep sill (77°18’N, 30°E) at the location of the Polar Front between AW and PW. The ship transects are complemented by five underwater glider missions, two equipped with microstructure sensors. On the sill, we observe warm (>2°C) and salty (>34.8) AW intruding below the colder (<0°C) and fresher (34.4) PW setting up a geostrophic balance where currents exceed 20 cm/s. We observe anomalous warm and cold-water patches on the cold and warm side of the front, respectively, collocated with enhanced turbulence, where dissipation rates range between 10<sup>-8</sup> and 10<sup>-7</sup> W/kg. In addition, tidal currents on the sill reach 15 cm/s. The variable currents affect the front structure differently in the vertical. While the mid-depth location of the front is shifted by several kilometers, the location of the front near the bottom remains stationary.  The frontal dynamics on the sill result in transformation and mixing of AW, manifested in the troughs north of the sill as modified AW.</p>

<p>Kongsfjorden is an Arctic fjord situated in the Svalbard archipelago. The fjord hydrography is influenced by the warm and saline Atlantic water from West Spitsbergen Current (WSC) flowing northward on the shelf slope and the cold and fresh polar waters circulating on the shelf. Once several conditions are satisfied, Atlantic waters from the WSC can extensively flood the fjord. These intrusions are majorly confined to the summer season although some strong events have been identified also in winter. In this study, a decade of continuous observations is used to examine changes in water properties and water masses variability in inner Kongsfjorden, with a special focus on Atlantic water intrusions. Data have been gathered by the National Research Council of Italy (CNR) through the Mooring Dirigibile Italia (MDI) in addition to summer CTD surveys. MDI was deployed in September 2010 at 100m depth and comprises various temperature and salinity sensors placed at different depths. Analysis of the longest temperature series reveals a positive linear trend since 2010. However, both temperature and salinity present a peak at the beginning of 2017 and decreasing values toward the end of the series. No significant trends were found when considering the monthly water column temperature as average of few sensors’ measurements. Yet, differentiating the seasonal contributions reveals that summer temperatures feature a fast warming (0.26 °C/yr) whereas winters do not show a statistically significant linear trend. Temperature and salinity observations gathered at 25 and 85m depth are used to depict water masses variability accorfing to previous water masses classifications. Some evidences are noted: first, events of Atlantic water intrusions are always confined near the bottom and they are never seen at 25m, whilst summer freshwater is found only in the near surface. Second, the timing of occurrence of these two water types seems to be related: the presence of large freshwater volumes close to the surface are preceded by the arrival of warm and saline waters. This evidence is interpreted as the melting signal of Kronebreen, the largest tidewater glacier in Kongsfjorden, triggered by the intrusion of Atlantic water. As a result, the large freshwater input manages to dilute the Atlantic water settled near the bottom. Third, the temperature and salinity peaks at the beginning of 2017 are associated to a massive Atlantic water flooding in the inner fjord lasting several months in summer/autumn 2016 and 2017. After this period, Atlantic water is seen only for few months in summer 2019. CTD measurements are used to depict the summer hydrography of Kongsfjorden and the focus is drawn on the characterisation of the seasonal cycle for each available year of measurements. Finally, the drivers of Atlantic water intrusion events are examined, as the presence of low pressure systems and the wind patterns in the region.</p>

Most of the exchange of water, salt and heat between the Arctic Mediterranean and the worlds oceans occurs through the Framstrait and the Greenland Sea. Our present knowledge on the respective northward and southward water mass transports is essentially based on current meter moorings, geostrophic calculations from hydrographic measurements and a variety of drifters. In order to explore the spatial velocity structure - horizontally on scales of eddies larger than 10 to 20 km and vertically in the order of 10 m - we have used a ship-mounted ADCP on several expeditions of RV Polarstern since 1990 to investigate the velocity field within the uppermost 400 m. These measurements provide snap-shots of the velocity field on scales not resolved by moorings, and they also serve as reference for the conversion of geostrophic into absolute velocities. Furthermore, in combination with water mass analyses it was possible to calculate the individual transports of the characteristic water masses for the whole water column in addition to the total transport in the area. The combination of high resolution hydrographic and velocity measurements at identical grid points allows to avoid the interpolation problems involved in the evaluation of mooring measurements.The mean circulation of the Greenland Sea is dominated by a large cyclonic and predominantly barotropic gyre. The calculated absolute velocities across the 75°N standard section question the existence of Koltermanns (1991) postulated deep anticyclonic gyre. At 75°N the East Greenland Current (EGC) is identified over a distance of 140 km as a narrow jet which carries ice and polar water to the South. The total volume transport calculated for the region of the EGC is comparable with results of moored current meters and ranges between 12 and 29 Sv (Fahrbach et al., 1995 and Woodgate et al., 1999).In contrast to the EGC the Westspitsbergen Current (WSC) carries Atlantic Water (AW) to the North and exhibits a much larger mesoscale variability. The velocity field in the WSC is characterized by variable meanders and mesoscale eddies with typical horizontal dimensions of 50 km, whereas jet-like structures dominate in the EGC. Since the AW provides the major contributions to the meridional heat transport five realizations of the 75°N standard section were used to investigate its interannual variability. During the summers 1990 - 1998 the AW transports ranged between 2 and 7 Sv. The total heat transport across 75°N is estimated as 52 TW in September 1997 and 42 TW in September 1998. The total salt transport ranges between 5.2 and 5.6 × 106 kg s-1. Finally, based upon five hydrographic sections taken between 70°N - 82°N and 25°W - 25°E during August/September 1997 a circulation and transport scheme for the principle water masses is constructed and compared to the results of Mauritzens inverse box model (Mauritzen, 1994). Both transport schemes are in good agreement. Between 75°N and 79°40N the mean temperature of the AW decreases by 0.8°C while its density increases. The observed AW cooling is caused by a strong heat loss to the atmosphere of about 130 W m-2.

The covariability of the Atlantic Water (AW) branches in the Nordic Seas is investigated over the period 1979–2012 using an eddy permitting model. A noticeable circulation change is found in the mid‐1990s. Prior to the mid‐1990s, the leading mode of variability defines a large‐scale pattern, with concomitant variations in the Atlantic Water (AW) inflow in the Faroe‐Shetland Channel (FSC), the Norwegian Atlantic Slope Current (NwASC), the AW inflow to the Barents Sea, and the West Spitsbergen Current (WSC). After the mid‐1990s, the covariability between the NwASC and the AW inflow in both the FSC and the WSC is lost. Consequently, the northern Barents Sea circulation anomaly pattern, which is present throughout the full period, becomes the leading mode of circulation in the northern Nordic Seas after the mid‐1990s. The circulation change of the mid‐1990s appears to be linked to a weakening of the southwesterly wind anomalies in Norwegian Sea, as the northern center of action of the first mode of sea level pressure weakens. Passive tracer experiments suggest that this circulation change may be accompanied by increased heat transfer from the AW current to the interior Nordic Seas. This in turn may have limited the influence of the recently observed AW warming in the Iceland‐Scotland Passage on the NwASC downstream.

Contrasting optical properties of surface waters across the Fram Strait and its potential biological implications

A major source of heat (and salt) for the Arctic Ocean is the Atlantic Water (AW) imported from the Norwegian Sea by the West Spitsbergen Current (WSC). Analysis of temperature records from the WSC has helped to link the warming of the Arctic Ocean to changes in the North Atlantic Oscillation (NAO) index. Here we analyze, using summer hydrographic data from a meridional section at 15°E in the northern Norwegian Sea, the interannual variability in the AW hydrography and its relation to the NAO index in the 1990s. We show that while the AW temperature exhibited a tendency to lag the NAO index, the AW salinity exhibited a tendency to lead the index. A surprising conclusion is that the AW temperature in the WSC was predictable one year in advance from the AW salinity.

As part of the N-ICE2015 campaign, IAOOS (Ice Atmosphere Ocean Observing System) platforms gathered intensive winter data at the entrance of Atlantic Water (AW) inflow to the Arctic Ocean north of Svalbard. These data are used to examine the performance of the 1/12° resolution Mercator Ocean global operational ice/ocean model in the marginal ice zone north of Svalbard. Modeled sea-ice extent, ocean heat fluxes, mixed layer depths, and AW mass characteristics are in good agreement with observations. Model outputs are then used to put the observations in a larger spatial and temporal context. Model outputs show that AW pathways over and around the Yermak Plateau differ in winter from summer. In winter, the large AW volume transport of the West Spitsbergen Current (WSC) (∼4 Sv) proceeds to the North East through 3 branches: the Svalbard Branch (∼0.5 Sv) along the northern shelf break of Svalbard, the Yermak Branch (∼1.1 Sv) along the western slope of the Yermak Plateau and the Yermak Pass Branch (∼2.0 Sv) through a pass in the Yermak Plateau at 80.8°N. In summer, the AW transport in the WSC is smaller (∼2 Sv) and there is no transport through the Yermak Pass. Although only eddy-permitting in the area, the model suggests an important mesoscale activity throughout the AW flow. The large differences in ice extent between winters 2015 and 2016 follow very distinct atmospheric and oceanic conditions in the preceding summer and autumn seasons. Convection-induced upward heat fluxes maintained the area free of ice in winter 2016. This article is protected by copyright. All rights reserved.

Submarine melting is an important balance term for tidewater glaciers1,2 and recent observations point to a change in the submarine melt rate as a potential trigger for the widespread acceleration of outlet glaciers in Greenland3-5. Our understanding of the dynamics involved, and hence our ability to interpret past and predict future variability of the Greenland Ice Sheet, however, is severely impeded by the lack of measurements at the ice/ocean interface. To fill this gap, attempts to quantify the submarine melt rate and its variability have relied on a paradigm developed for tidewater glaciers terminating in fjords with shallow sills. In this case, the fjords’ waters are mostly homogeneous and the heat transport to the terminus, and hence the melt rate, is controlled by a single overturning cell in which glacially modified water upwells at the ice edge, driving an inflow at depth and a fresh outflow at the surface1. Greenland’s fjords, however, have deep sills which allow both cold, fresh Arctic and warm, salty Atlantic waters, circulating around Greenland, to reach the ice sheet margin3,6,7. Thus, Greenland’s glaciers flow into strongly stratified fjords and the generic tidewater glacier paradigm is not applicable. Here, using new summer data collected at the margins of Helheim Glacier, East Greenland, we show that melting is driven by both Atlantic and Arctic waters and that the circulation at the ice edge is organized in multiple, overturning cells that arise from their different properties. Multiple cells with different characteristics are also observed in winter, when glacial run off is at a minimum and there is little surface outflow. These results indicate that stratification in the fjord waters has a profound impact on the melting dynamics and suggest that the shape and stability of Greenland’s glaciers are strongly influenced by layering and variability in the Arctic and Atlantic waters.

Fram Strait, located between Svalbard and Greenland, serves as a crucial gateway connecting the Arctic Ocean and the North Atlantic, facilitating the exchange of heat and freshwater between these regions. Warm and saline Atlantic Water (AW) is carried northwards by the West Spitsbergen Current (WSC), and constitutes the main source of oceanic heat and salt entering the Arctic Ocean. Variations in the AW inflow strongly influence both Arctic ocean and sea ice conditions. An array of moorings has been monitoring the year-round inflow of AW in the WSC, providing hydrographic and current data from 1997 – 2024. A robust, long-term AW warming trend of 0.20°C/decade is identified, leading to a total increase of 0.54°C over the 27-year record. Distinct multi-annual warm and cold anomalies are identified, lasting ~2 years, with two warm periods (2005–2007 and 2015–2017) and two cold periods (1997–1999 and 2019–2024), linked to distinct shifts in the AW temperature regime. Notably, the most recent cold anomaly persisted for over five years—more than twice the duration of previous events. The interannual variability in AW temperatures results from a combination of advection from upstream in the Nordic Seas and local atmospheric forcing. Temperature anomalies propagate northward into the Arctic Ocean along the AW inflow pathway to the north of Svalbard, with a 2-month lag relative to Fram Strait, thus the expected continued rise in AW temperatures and associated heat transport will have profound and lasting impacts on the future state of the Arctic Ocean.

<p>The interaction between North Atlantic and Arctic Ocean waters plays a key role in climate variability and in<br>driving the global thermohaline circulation. In the past decades, an increased heat input to the Arctic has<br>occurred which is considered of high climatic relevance as, e.g., it contributes to enhancing sea ice melting.<br>In this frame, the progressive northward extension of the Atlantic signal within the Arctic domain known as<br>Arctic Atlantification is one of the most dramatic environmental local changes of the last decades.<br>In this study we used in situ data and the Copernicus Marine Environment Monitoring Service (CMEMS)<br>reanalysis dataset to explore spatial and temporal variability of water masses on different time-scales and<br>depths in the eastern Fram Strait. In that area, warm and salty Atlantic Water (AW) enters the Arctic Ocean<br>through the West Spitsbergen Current (WSC). Time series of potential temperature, salinity and potential<br>density obtained from CMEMS reanalysis in the surface, upper-intermediate and deep layers referring to the<br>period 1991-2019 have been considered. High-frequency observations gathered from an oceanographic<br>mooring maintained by the National Institute of Oceanography and Applied Geophysics (OGS) in<br>collaboration with the Italian National Research Council - Institute of Polar Science (CNR-ISP) have been<br>used to assess the reliability of CMEMS data in reproducing ocean dynamics in the deep layer (ca 900-1000<br>m depth) of the SW offshore Svalbard area. The mooring system has been collecting data since June 2014.<br>In this contribution, we will show how the CMEMS data compared with in situ measurements as far as<br>seasonal and interannual variations as well as long-term trends are concerned. We will also discuss how<br>CMEMS reanalyses show differences in resolving ocean dynamics at different depths. Particularly, the severe<br>limitations in reproducing thermohaline variability at depths greater than 700 m. Finally, we will illustrate how<br>our results highlight strengths and limitations of CMEMS reanalyses, underscoring the importance of<br>optimizing measurements in a strategic area for studying climate change impacts in the Arctic and sub-Arctic<br>regions.</p>

Interannual variability of the Atlantic water layer in the West Spitsbergen Current at [formula omitted] in summer 1991–2003