Does the East Greenland Current exist in northern Fram Strait?

Warm Atlantic Water (AW) flows around the Nordic Seas in a cyclonic boundary current loop. Some AW enters the Arctic Ocean where it is transformed to Arctic Atlantic Water (AAW) before exiting through Fram Strait. There the AAW is joined by recirculating AW. Here we present the first summer synoptic study targeted at resolving this confluence in Fram Strait which forms the East Greenland Current (EGC). Absolute geostrophic velocities and hydrography from observations in 2016, including four sections crossing the east Greenland shelfbreak, are compared to output from an eddy-resolving configuration of 5 the sea-ice ocean model FESOM. Far offshore (120 km at 80.8◦ N) AW warmer than 2◦ C is found in northern Fram Strait. The Arctic Ocean outflow there is broad and barotropic, but gets narrower and more baroclinic toward the south as recirculating AW increases the cross-shelfbreak density gradient. This barotropic to baroclinic transition appears to form the well-known EGC boundary current flowing along the shelfbreak further south where it has been previously described. In this realization, between 80.2◦ N and 76.5◦ N, the southward transport along the east Greenland shelfbreak increases from roughly 1 Sv to about 4 Sv 10 and the warm water composition, defined as the fraction of AW of the sum of AW and AAW (AW/(AW+AAW)), changes from 19±8% to 80±3%. Consequently, in southern Fram Strait, AW can propagate into Norske Trough on the east Greenland shelf and reach the large marine terminating glaciers there. High instantaneous variability observed in both the synoptic data and the model output is attributed to eddies, the representation of which is crucial as they mediate the westward transport of AW in the recirculation and thus structure the confluence forming the EGC. 15

rim south of 81 • N (Rudels et al., 2005). There is evidence from drifter data (Gascard et al., 1995), hydrographic surveys (Marnela et al., 2013), an inverse modelling study (Schlichtholz and Houssais, 1999) and a numerical ocean model (Kawasaki and Hasumi, 2016) that the recirculation in Fram Strait may extend beyond 81 • N, possibly as far north as 82 • N. However, evidence from model studies in Fram Strait is at present inconclusive as the northern limit of the recirculation, the strength of individual recirculation pathways and of the boundary currents varies between models (Maslowski et al., 2004;Aksenov 5 et al., 2010;Hattermann et al., 2016;Ilicak et al., 2016;Wekerle et al., 2017). This appears to be related to the resolution of the models and the bathymetry (Fieg et al., 2010). Observations able to determine the strength and location of the recirculation are therefore needed.
However, due to heavy sea-ice conditions observational studies in the central and western Fram Strait significantly north of 79 • N are scarce. Thus, the northern AW recirculation and EGC remain undersampled and poorly understood. A study of 10 the Arctic Ocean outflow along the northeast Greenland shelfbreak using data from 82-83 • N (Falck et al., 2005) shows no recirculating AW there. South of 79 • N, the EGC is a current located offshore of the Greenland shelfbreak on the western side of Fram Strait that transports recirculated AW and modified AAW below relatively fresh and cold PSW and sea-ice from the Arctic (see Aagaard and Coachman, 1968, for a review of early observations of the EGC). Both recirculating AW and AAW lose contact with the atmosphere before reaching northern Fram Strait. The different transit times through the Arctic Ocean

15
(~1 year to 10s of years: Karcher et al., 2003;Polyakov et al., 2011) compared to the recirculation in Fram Strait (on the order of months for AW: Gascard et al., 1995;Hattermann et al., 2016) have also been inferred from the lower oxygen saturation of AAW compared to AW. Between 78 • N and Denmark Strait the EGC consists of three branches: an inshore branch transporting fresh cold water, a shelfbreak branch and a branch offshore of the shelfbreak believed to be a direct recirculation of AW from the western WSC branch (Håvik et al., 2017;Nilsson et al., 2008;Woodgate et al., 1999). 20 The aim of this study is to utilize the first synoptic data set targeted at investigating the structure of the AW recirculation and of the EGC in Fram Strait north of 79 • N. We will describe the hydrography (potential temperature, salinity, potential density) and the kinematics (absolute geostrophic velocity fields) along the path of Atlantic Water (AW) in Fram Strait. We start with the inflow of AW and the WSC at 79 • N in Sect. 3.1. Then, we turn to central Fram Strait and the westward recirculation of AW crossing 0 • EW (Sect. 3.2) before we follow the path of the southward flow along the east Greenland shelf from~80.3 • N 25 to 76.6 • N in Sect. 3.3. We will examine the formation and transport of the EGC in Sect. 4.1 and take a look at shelf processes on the northeast Greenland shelf in Sect. 4.2. Throughout, we utilize an eddy resolving numerical model (Wekerle et al., 2017) to put the synoptic observations in a larger temporal and spatial context and to assess in which state of the highly variable flow regime the observations were taken. We close with conclusions from our findings in Sect. 5.
2 Data and methods 30

CTD and ADCP data
Data was collected between 18 th of July and 6 th of September 2016 during cruise PS100 of RV Polarstern. The data consists of 75 stations along 5 sections (0 • EW, 79 • N, WT1, 79.6 • N and NT1; Fig. 1). CTD casts (Kanzow et al., 2017a, b) were recorded with a dual duct Seabird 911+ and averaged into 1 m bins (Kanzow, 2017). The conductivity and oxygen sensors were calibrated using water samples analysed on-board (Kanzow, 2017) with an Optimare Precision Salinometer and with a titration method, respectively. An upward and a downward looking 300 kHz RDI Workhorse Acoustic Doppler Current Profiler (ADCP) were used as a lowered-ADCP (LADCP) system . A vessel-mounted 150 kHz RDI Ocean Surveyor ADCP (VMADCP) recorded ocean velocities along the cruise track (Kanzow and Witte, 2016). ADCP velocities 5 were detided by subtracting the barotropic tidal component calculated from the Arctic Ocean Tidal Inverse Model (AOTIM-5, Padman and Erofeeva, 2004). VMADCP and LADCP setup and processing are described in detail in Kanzow (2017).

Data processing
For each section, station locations are projected onto the straight lines shown in Fig. 1 retaining their longitude (latitude in the case of 0 • EW). Bathymetry information from the ship's echosounder, the IBCAO V3 bathymetry (Jakobsson et al., 2012) and 10 CTD altimeter station depths agreed to within 10s of meters. Therefore, we use the linearly interpolated station depths to plot the bathymetry in the sections. In section WT1 the location of the shelfbreak is corrected using the echosounder bathymetry.
The easternmost bathymetry at 79 • N near the Svalbard shelf is corrected using IBCAO bathymetry of the Svalbard shelfbreak.
In section 0 • EW we use the bathymetry from IBCAO for the entire section and interpolated hydrographical values appearing below the so-defined seafloor are removed before plotting. 15 For each CTD station the VMADCP velocity profiles are averaged whilst the ship was on station to attain a single profile. For each section the station data (CTD, LADCP and VMADCP) are interpolated onto a common grid with vertical resolution of 10 m and a horizontal resolution of half the mean station distance of the section (ranging from 5 to 20 km) using a Laplacian-Spline interpolation (Smith and Wessel, 1990). A standard tension of 5 (0 = Laplacian interpolation, ∞ = spline interpolation) and a search radius of 10 grid points are used. Geostrophic shear is calculated from the gridded hydrography using thermal 20 wind and is referenced to the 50-150 m averaged on-station VMADCP velocities (except for section NT1 where the 50-150 m LADCP data is used) to obtain absolute geostrophic velocities. For conceptual considerations, we additionally use a simple two layer ocean approximation with a density difference of 0.3 kg m −3 to estimate baroclinic velocities from the slope of the 27.8 kg m −3 isopycnal. The position, width and core velocity of the shelfbreak EGC and WSC are defined following Håvik et al. (2017): The core velocity is the maximum of the 0-150 m mean velocity of the section. The boundaries of the EGC and 25 WSC are defined as the locations where the 0-150 m mean velocity has decreased to 20 % of the core value. This criterion is also used to define the boundaries of the EGC within which we calculate net transport. It has the advantage over using a fixed width or distance from the shelfbreak that it can account for a meandering current of variable width, as we expect to see in synoptic observations.
To assess the errors due to the gridding process, the CTD and ADCP data are regridded increasing or decreasing a) tension, 30 b) search radius and c) grid resolution individually by a factor of two. The relative absolute error of the absolute geostrophic velocity between the modified grid and the grid used in this study is determined. Velocity error estimates from a and b are generally below 10 %, with some higher values occurring below 500 m outside of the EGC at 79.6 • N. Velocity errors from c are mostly below 30 %, higher values are found in areas of large and uneven station spacing. Note that a change in grid spacing of factor two is rather large and thus presents a maximum error estimate. The error of the VMADCP measurements is calculated as the median absolute deviation over the full sampling depth in time and space whilst on station and is~0.04 m s −1 with maximum values of 0.07 m s −1 . The processing routine for LADCP velocities gives an error estimate dependant on depth for each cast (Thurnherr, 2010;Kanzow, 2017). The median error between 50 and 150 m depth at section NT1 is below 0.05 m s −1 for all except for the eastern most station where it is 0.1 m s −1 .

Numerical model
In this study we use model output from the Finite-Element Sea-ice Ocean Model (FESOM) version 1.4 (Wang et al., 2014;Danilov et al., 2015). FESOM is an ocean-sea ice model which solves the hydrostatic primitive equations in the Boussinesq approximation. The sea ice component applies the elastic-viscous-plastic rheology (Hunke and Dukowicz, 2001) and thermodynamics following Parkinson and Washington (1979). The finite element method is used to discretise the governing equations, 10 applying unstructured triangular meshes in the horizontal and z-levels in the vertical.
We use a global FESOM configuration that was optimised for Fram Strait, applying a mesh resolution of 1 km in the area 75 • N-82.5 • N/20 • W-20 • E and 4.5 km in the Nordic Seas and Arctic Ocean (Wekerle et al., 2017). In comparison to the local Rossby radius of deformation (around 4-6 km in Fram Strait, von Appen et al., 2016), this configuration can be considered as "eddy-resolving". The model bathymetry was taken from RTopo2 (Schaffer et al., 2016). The model is forced with the 15 atmospheric reanalysis data COREv.2 (Large and Yeager, 2009), and interannual monthly river runoff is taken from Dai et al. Comparison with various observational data showed that the model generally performs very well in terms of circulation structure, eddy activity and hydrography (Wekerle et al., 2017) which makes us confident that we can use it as a best-estimate 20 realistic hindcast of the circulation and hydrography in Fram Strait. However, there is a bias toward higher salinity in the Atlantic Water layer of around 0.15. This salinity bias can be traced back into the North Atlantic, and is a result of model deficiencies in correctly representing the pathways of the North Atlantic Current.
Eddy kinetic energy (EKE) is computed by decomposing velocities u and v into monthly means (denoted by bar) and a deviating part (denoted by prime). The time-averaged EKE is then

Watermass definitions and calculations
Watermass definitions (see Table 1) follow Rudels et al. (2005)  defined as the depth of maximum salinity (Richter, 2017). Endmembers for mixing calculations are picked as the deepest water sampled (DW), the warmest subsurface θ-S peak found in the AW inflow region at 79 • N (AW), the coldest clearly defined deep temperature maximum (AAW) and the coldest water sampled (PSW) and are given in Table 2. Since AW and/or AAW is always located between PSW and DW, and since DW and PSW are not observed to mix, we can describe our observations as either AW-AAW-PSW mixtures or as AW-AAW-DW mixtures. The resulting mixing triangles are shown in Fig. 2. Note 5 that the relative contribution of AW and AAW in a water parcel that is mostly comprised of AW and AAW is not affected by this method. Errorbars for the watermass fractions are calculated by repeating the calculation 1000 times including random normally distributed uncertainties for the temperature and salinity of the endmembers with a standard deviation of 0.2 • C and 0.04 PSU respectively. Please note that the distribution of uncertainties naturally includes values outside of the ±1 standard deviation boundary. The reported uncertainties correspond to the standard deviation over all realizations of the watermass 10 calculation.

Results
We now present our results following the path of Atlantic Water through Fram Strait, from the inflow in the WSC, via the recirculation in central Fram Strait to the EGC. A particular emphasis is placed on the formation and evolution of the EGC.

15
The most striking feature of section 79 • N, as measured in summer 2016, is the highly dynamic velocity field (Fig. 3c). This of 0.11 m s −1 (Fig. 3c). We did not observe an offshore branch of the WSC which is consistent with long term measurements where the offshore branch is observed to be weakest or absent during summer months (von Appen et al., 2016;Beszczynska-Möller et al., 2012).
The water column in the WSC is temperature stratified with a temperature maximum at the surface, while the minimum temperature is in the deep ocean (Fig. 3a+b). The surface temperatures of over 9 • C on the west Spitsbergen slope are the 5 highest water temperatures in the WSC near 79 • N published so far and are likely due to the warming of the AW inflow to Fram Strait (Beszczynska-Möller et al., 2012;Walczowski et al., 2017). The AW layer is over 500 m thick and is in contact with the atmosphere east of 5 • E (Fig. 3a). Toward the west the AW layer gets thinner and the depth of the temperature maximum increases. Although water warmer than 2 • C is found in the upper 50 m west of 5 • E, this water is too fresh to fall into the AW definition ( Fig. 3a/b).

The westward recirculation in the deep Fram Strait
The synoptic section in central Fram Strait shows a south to north transition along 0 • EW. At the southern most station (near 78 • N) the water has an almost uniform salinity with warm AW close to the surface (the water in the upper tens of meters is too fresh to fall into the AW definition) and colder water at depth (Fig. 5a), similar to the stations sampled in the WSC along 79 • N (Fig. 3a). With increasing latitude the observed AW layer gets thinner, colder, fresher and is located deeper in the water 15 column. This means that between the AW inflow at the surface in the WSC and the subsurface AW layer in the northern part of the central Fram Strait, AW has to subduct underneath colder and fresher PSW and sea-ice. This was also simulated in the eddy resolving model study of Fram Strait by Hattermann et al. (2016) and it was hypothesized that baroclinic instability may achieve this subduction. The subduction of AW under PSW is also simulated in FESOM though this does not show a northward thinning of the AW layer ( Fig. 5b). In the observations the Arctic Ocean halocline, with cold, fresh PSW at the surface, is found 20 in the upper 120 m of the water column north of 80 • N below which Knee Water (KW, the saltiest water close to the freezing point line) is found. The properties of KW are indicative of the ice-ocean-atmosphere interaction in the Arctic Ocean (Moore and Wallace, 1988;Rudels et al., 2005) signalling that we observe water modified in the the Arctic Ocean north of 80 • N. In addition to their maximum temperature (more or less than 2 • C) AW and AAW along 0 • EW exhibit differences in oxygen saturation. Since AAW has transited through the Arctic Ocean, its oxygen saturation of typically~80 % is significantly lower 25 than the oxygen saturation of AW of typically~100 %.
AW is present somewhere in the watercolumn at all stations along 0 • EW except for the northernmost station at 80.8 • N ( Fig. 5a). This implies that we sampled either the northern rim of the recirculation as it was at the time of our measurements, or that we sampled a passing AAW filament. We cannot decide which of the two explanations is true since no measurements further north than 80.8 • N were taken during the cruise. Examining the mean temperature in FESOM at 0 • EW ( Hence the model does not allow us to judge which of the two possible explanations is more likely. The synoptic observations made here do however show that the recirculation in Fram Strait can reach as far north as 80.7 • N. A repeat synoptic survey along 0 • EW, with a higher resolution than in the present study, extending beyond 81 • N, supported by a mooring array, could provide a more definite picture of the northern limit of Fram Strait recirculation and its meridional and temporal structure. This 5 could then be used to verify numerical models. In the central Fram Strait along 0 • EW (Fig. 5c) the measured absolute geostrophic velocity field switches between broad sectors of weak eastward (~78 • N and~79.5 • N) and westward velocity (~78.5 to~79 • N and around 80 to 80.5 • N). Velocities reach ± 0.12 m s −1 . The velocity field is mostly barotropic but the station spacing of~40 km is not able to resolve the flow structure. We expect the velocity field, at least in the vicinity of 79 • N, to be similar to the velocity field shown in Fig

The evolution of the EGC from northern Fram Strait to the Greenland Sea
In the synoptic section roughly perpendicular to the east Greenland shelfbreak at~80.3 • N (Section WT1), AW is only found in the central Fram Strait near 0 • EW, some 130 km east of the Greenland shelfbreak (Fig. 3a). This is closer to the Svalbard 25 shelfbreak than the Greenland shelfbreak. The deep θ-S maxima sampled west of 0 • EW at WT1 have temperatures around 1 • C, well below the temperature of AW, and salinities between 34.8 and 34.9 (Fig. 6). This agrees with deep θ-S maxima from stations sampled between 82-83 • N and 10-5 • W in 2004 (Rudels et al., 2012), which, together with the transport measured there (Marnela et al., 2008) indicates that the AAW sampled at WT1 may be advected from the northwest along the east Greenland shelfbreak. Thus, the Arctic Ocean outflow of AAW sampled at 80.3 • N is uninfluenced by directly recirculating 30 AW west of 0 • EW. Salinity (Fig. 3b) increases strongly in the halocline over the upper 150 m. The density field (thin contour lines in Fig. 3b) closely follows the salinity field. At the mouth of Westwind Trough the temperature of the deep θ-S maximum is~0.8 • C.
Outside of the trough, two regions of southward flow were sampled (Fig. 3c). The local velocity maximum between 0-20 km offshore of the shelfbreak with relatively weak core velocities of -0.09 m s −1 is at a cross-shelfbreak distance where the shelfbreak EGC is found further south. The broad southward flow between 5 • W and 0 • EW (30 km and 120 km), identified as the Arctic Ocean outflow, is also visible in the modeld velocity field (Fig. 4b). Both bands of southward flow are highly barotropic and modelled EKE is negligible at WT1 (Fig. 4c).
In the Arctic Ocean outflow, at~80.3 • N (section WT1), the slope of the 27.8 kg m −3 isopycnal between 0 • EW and the Thus we hypothesise that the southward flow at WT1 may not be a boundary current tied to the shelfbreak.
In the eight-year model average the AW reaches much closer to the shelfbreak at 80.3 • N than in the synoptic section (  It may thus impact the distance from the shelfbreak at which AW is found in the model. From comparison with the sparse observations available (this study, a synoptic section in Rudels et al. (2005) and the climatology in Schaffer et al. (2017)) we are inclined to trust the density and velocity field in FESOM in northern Fram Strait, but are more cautious about the distribution of AW. Thus, correctly modelled currents may advect the wrong water mass in the model, specifically AW may be simulated too far in the west. 20 Just 50 km further to the south, at 79.6 • N, AW is found merely 30 km offshore of the shelfbreak in a core between 150-450 m depth (Fig. 3a). The 27.8 kg m −3 isopycnal has a downward slope of 0.5 m km −1 toward the shelfbreak (this corresponds to a baroclinic velocity of 0.1 m s −1 ), which has a greater similarity to the EGC structure further south (Håvik et al., 2017) than the WT1 section. The offshore divergence of the isopycnals may be caused by AW intruding below, into and/or above the AAW layer at depth. The spreading apart of the isopycnals in the ambient AAW by intruding AW is likely a generic process 25 (i.e. not just present in this synoptic section), taking place whenever AW meets AAW at depth with a distinct and strong horizontal gradient in stratification. Intruding AW at depth has lower stratification consistent with the strong atmospheric cooling experienced relatively recently by the AW in the Nordic Seas boundary current loop. Some interleaving is present in the CTD profiles at the transition between AW and AAW 30 km from the shelfbreak (orange profile in Fig. 6). Largely barotropic southward velocities (~0.16 m s −1 , Fig. 3c) are found just offshore of the shelfbreak. EGC was a boundary current and close to the shelfbreak. The latter is supported by the fact that in the model the southward flow at 79.6 • N lies closer to the shelfbreak in summer than in winter (Fig. 8). Conversely, the upward sloping isopycnals seen below 200 m suggest the presence of an AW eddy in the synoptic section.
Another 80 km further to the south, at 79 • N, AW is found at~200 m depth at the east Greenland shelfbreak though no 5 AW is found on the east Greenland shelf (Fig. 3a). The 27.9 kg m −3 isopycnal undulates strongly, following the temperature At 79 • N there are two cores of southward velocities (Fig. 3c). We identify the core just offshore of the shelfbreak centred around 5 • W (20 km) and reaching -0.15 m s −1 as the shelfbreak EGC.
The modelled average temperature and velocity field are naturally smoother than the synoptic section but show the same general 15 structure with AW subducting westward below PSW (Fig. 4a). The EKE at 79 • N is much higher than at the sections sampled to the north and south of this and has a peak where the EGC is found. This high variability can also be seen in the daily averages of the velocity field (Supplementary Material: Fig. S1).
At the mouth of Norske Trough (76.6 • N, i.e. another 270 km further to the south along the shelfbreak), AW is found in a broad core between 100 and 350 m depth at and offshore of the shelfbreak (Fig. 3a). Inside of the trough a thin layer of AW 20 is found between 200 and 250 m, i.e. above 320 m which is the depth of the shallowest sill between the shelfbreak and the inner shelf near the NEGIS glaciers . The model also shows an AW layer within Norske Trough, both in the eight-year average (Fig. 4a) and in the daily averages for 2009 (Supplementary Material: Fig. S1). Thus, AW is able to propagate through Norske Trough to the termini of the NEGIS glaciers. However, the modelled AW layer is thicker inside Norske Trough than in the observations and thins eastward. Since this does also not agree with the temperature observations in 25 Norske Trough reported in , we again conclude that the model transports too much AW too far eastward.
The temperature of the synoptic deep θ-S maximum decreases from east to west and its depth increases (Fig. 6). Observed salinities (Fig. 3b) are lowest at the surface and on the shelf. The density field largely follows the salinity field and isopycnals deepen toward the west (Fig. 3b). The 27.8 kg m −3 isopycnal has a downward slope of 1.66 m km −1 toward the west which corresponds to a baroclinic velocity of 0.33 m s −1 . The location and width of the shelfbreak EGC at NT1 agree well with Section 10 from Håvik et al. (2017), which is located 30 km to the north of section NT1.

Discussion
In the following we will examine the evolution of the Arctic Ocean outflow to the EGC from north to south. The change in dynamics is addressed by examining the baroclinic and barotropic components of the southward flow. We will then discuss the 5 transport along the shelfbreak, examining the different watermasses (e.g. DSOW transport) and compare this with observations of the EGC further south. Finally we will draw inferences from our results about the circulation on the northeast Greenland shelf.

Formation of and transport in the EGC
Both the observations and the model indicate that the recirculating AW first gets close to the east Greenland shelfbreak between 10 the mouth of Westwind Trough at 80.3 • N and 79.6 • N. From our observations (Fig. 3) and the modelled velocity field of the AW layer in Fram Strait (Fig. 8c+d) we argue that this is likely to take place closer to 79.6 • N than to WT1. The three sections crossing the EGC downstream of WT1 show different stages of watermass transformation in the deep temperature maximum ( Fig. 6): from AAW and AW located horizontally next to another at 79.6 • N to successively greater mixing between the two until the deep temperature maximum is warmer than 2 • C (and thus falls into the AW definition) at all stations sampled in 15 section NT1. Successively more AW gets entrained into the core of the EGC from north to south, with the contribution of the AW endmember increasing from only 19±8 % at WT1 to 80±3 % at NT1 (Fig. 7c). This can also be seen in the north to south increase of temperature, salinity and oxygen concentration, and the decrease of the depth and density of the deep temperature maximum within the core of the southward flow ( Fig. 7a The transport of the southward flow along the shelfbreak varies between -0.9 Sv at WT1 and -4.0 Sv at NT1 and generally increases downstream (Fig. 7d). The exception is 79 • N where the shelfbreak EGC transports only 1.1 Sv which is over 1.5 Sv to -0.26 m s −1 at NT1, again with section 79 • N an exception (Fig. 7d). The transport through 79 • N is also low when compared with previous estimates of southward transport through 79 • N (e.g. Fahrbach et al., 2001;de Steur et al., , 2014Schlichtholz and Houssais, 1999). Between 79 • N and 78 • 50' N the summer mean EGC transport increases by~2 Sv (de Steur et al., 2014) implying that recirculation of this magnitude joins the EGC between these two sections (this transport estimate includes, but is not restricted to AW). In winter, the transport increases by an additional~3 Sv between the two latitudes, likely 5 due to an intensification of the Greenland Sea Gyre (de Steur et al., 2014). Even the summer increase of 2 Sv is higher than the increase in our synoptic summer transport between 79.6 • N and section NT1 (Fig. 7d). our study is synoptic. It also has to be kept in mind that the station spacing of the moorings is wider than our station spacing 10 and thus interpolation between moorings may remove much of the small scale variability that reduces transport in the synoptic section at 79 • N. Since the synoptic AW transport is the majority of the total synoptic EGC transport, "missing" transport would most likely be AAW and PSW. These watermasses are found on the shelf at 79 • N where a surface intensified jet, which appears to be similar to the PSW Jet of Håvik et al. (2017), has a southward transport of 1.1 Sv (Fig. 3c).
The denser components of both AW and AAW are > 27.8 kg m −3 which is the density definition of Denmark Strait Overflow 15 Water (DSOW) (Fig. 7d). Transport of DSOW increases from -0. separation of the EGC into multiple branches south of this latitude, the transport begins to decrease. The transport of the shelfbreak EGC of their Section 10 (for location see Fig. 1) agrees with our estimate for NT1 though our velocities are significantly lower (Fig. 7d). Velocities at their Section 9 were closer to our value for NT1 though transport and current width were higher.
Velocities and current widths measured by Håvik et al. (2017) were generally higher than those recorded in the present study. 30 This is consistent if one assumes that the increase in isopycnal slope seen between WT1 and NT1 (Sect. 3.3) continues further to the south. Another explanation could be the denser station spacing in Håvik et al. (2017) (5-7 km versus 10-20 km in our study). With a denser station spacing it is more likely to sample the location in the EGC with the highest velocity, thus making it more likely to arrive at a higher core velocity. Nevertheless, our study is able to extend the work by Håvik et al. (2017) northward of 79 • N.

Impact of the EGC on the northeast Greenland shelf
The depths of the deep temperature maximum, of the 1 • C isotherm inside Norske Trough and of the 0.5 • C isotherm inside is formed in the Arctic Ocean by ice-ocean-atmosphere interaction (e.g. Rudels et al., 2005). The distribution of KW is an important indication of the shelf circulation. Both at 79 • N (not shown) and inside Westwind Trough (Fig. 6), KW is markedly absent at stations close to and on the east Greenland shelf. At NT1 the situation is reversed: KW is only found inshore of the shelfbreak (Fig. 6). Observations close to the cavity of 79NG show that the KW signal found inside Norske Trough is eroded by isopycnal mixing with glacially modified water originating from both subglacial discharge and submarine melting 5 (Schaffer, 2017). This leads to the hypothesis that KW is brought to the glaciers via Norske Trough and waters without the something that should be investigated in the future is that water warmer than 2 • C spreads further northwestward in the model than observed there.
It is evident from the synoptic CTD sections along 79 • N in Marnela et al. (2013), in Langehaug and Falck (2012)  which make the velocity field appear rather smooth (Fig. 4b). In contrast, Fig. 3c shows a qualitative picture of instantaneous eddy variability. Considering the highly variable structure of the flow field is important to reconcile measurements that at first seem counter-intuitive, such as northward flow in areas where the EGC is expected, with the overall circulation in Fram Strait.
The daily velocity averages (Supplementary Material: Fig. S1) and long-term EKE averages from FESOM (Fig. 4c) show that the very dynamic velocity structure at 79 • N is not an artefact of our measurement technique. It rather is representative of 10 the synoptic eddy field, a view that is typically lost in depictions of long-term (or even monthly) averages. The same is true for the surface intensified current on the shelf (the PSW Jet) that is seen in the synoptic section. This, too, is not discernible in the multi-year average of the FESOM velocity field though it is sometimes present in the daily averages. We think that it is representative that in our synoptic sections the boundary currents (WSC and EGC) instantaneously appear weaker than the eddies present in Fram Strait. The synoptic view presented here is also important for understanding the manifold processes, 15 such as salt and heat transport to central Fram Strait and nutrient exchange between the surface layer and deeper watermasses, that are mediated by small scale features such as eddies. The small scale and highly variable structure of the velocity field in Fram Strait makes it essential to conduct both hydrographic surveys and model runs at an appropriate resolution to prevent aliasing. At the same time, it needs to be considered that any particular water sample taken in Fram Strait derives from this eddy field and may either have originated from inside or outside of transient eddies. 20 The Arctic Ocean outflow region between the northeast Greenland shelf and 0 • EW is evident as a broad barotropic flow both in our synoptic section at 80.3 • N (Fig. 3c) and in the velocity field from FESOM (Fig. 4b). From examination of the modelled velocity output we hypothesise that this Arctic Ocean outflow is at least partly topographically steered (see Fig. 7 in Wekerle et al., 2017). We propose that the evolution of the barotropic Arctic Ocean outflow to the baroclinic EGC is driven by the recirculation of AW in Fram Strait. As the recirculating AW reaches ever closer to the east Greenland shelfbreak, the 25 Arctic Ocean outflow is restricted to an increasingly narrow band along the shelfbreak. At the same time the density difference between recirculating AW and waters of Arctic origin drives a baroclinic current. In northern Fram Strait, where the maximum westward extent of AW is located in central Fram Strait close to 0 • EW, a baroclinic current associated with the Polar Front and the ice edge was described to merge with the EGC further south (Schlichtholz and Houssais, 1999). This current was described as part of the EGC by Paquette et al. (1985). Further south, the baroclinic boundary current EGC was also associated 30 with the Polar Front, there located at the east Greenland shelfbreak as recirculating AW has spread further west in Fram Strait.
Here we argue that the EGC, Arctic Ocean outflow and AW recirculation are not separate but that the latter two combine to form the EGC. In a more global perspective, there are other boundary currents which do not follow a shelfbreak in their upstream part; these have to join the shelfbreak somehow. For example, different idealized models (Lighthill, 1969;Endoh, 1973;Suginohara, 1980) showed that barotropic and baroclinic Rossby waves from the ocean interior can explain the formation of western boundary currents. Seemingly eddies may play the same role as Rossby waves. We further presume that eddies in Fram Strait transport warm water to the western boundary which increases the along boundary transport.
Aspects of the circulation that require further study are the northern extent of the recirculation, the spatial distribution of AW between the shelfbreak near Westwind Trough and 0 • EW, the circulation structure in the central Fram Strait north of 79 • N, with the possible role of the Molloy Hole, and the shelf circulation.

5
To answer the question whether the EGC exists in northern Fram Strait, we note that the baroclinic boundary current does not exist in northern Fram Strait. Here, we take southward flow in a baroclinic boundary current along the shelfbreak as a defining feature of the EGC. By this definition, based on our evidence, we conclude that the EGC does not exist north of 79 • N. It rather appears that the southward transport in northern Fram Strait is the Arctic Ocean outflow.
Author contributions. CW lead the analysis of the model output. MER lead the analysis of the data and interpretation of the data and model output, as well as the write up of the paper. All authors contributed to each of these points.
Competing interests. The authors declare that they have no conflict of interest.
Acknowledgements. We wish to thank the captain and crew of RV Polarstern. We would also like to thank Torsten Kanzow and Janin

15
Schaffer for their contributions to the comprehensive data set that this study is based on. Further we thank Janin Schaffer for the kind use of the map of Fram Strait in Fig. 1 and for useful discussions on glacier-trough exchange. Support for this study was provided by the           Figure 6. Potential temperature-salinity diagrams for three sections crossing the east Greenland shelfbreak (WT1, 79.6 • N and NT1). Individual casts are colour coded depending on their distance to the east Greenland shelfbreak (positive = offshore). Please note that the x-axis changes scale at 33. The solid black line shows the watermass boundary between AW and AAW (see Table 1).