Ice-Tethered Profilers (ITP), deployed in the Arctic Ocean between 2004 and 2013, have provided detailed temperature and salinity measurements of an assortment of halocline eddies. A total of 127 mesoscale eddies have been detected, 95% of which were anticyclones, the majority of which had anomalously cold cores. These cold-core anticyclonic eddies were observed in the Beaufort Gyre region (Canadian water eddies) and the vicinity of the Transpolar Drift Stream (Eurasian water eddies). An Arctic-wide calculation of the first baroclinic Rossby deformation radius R d has been made using ITP data coupled with climatology; R d 13 km in the Canadian water and 8 km in the Eurasian water. The observed eddies are found to have scales comparable to R d . Halocline eddies are in cyclogeostrophic balance and can be described by a Rankine vortex with maximum azimuthal speeds between 0.05 and 0.4 m/s. The relationship between radius and thickness for the eddies is consistent with adjustment to the ambient stratification. Eddies may be divided into four groups, each characterized by distinct core depths and core temperature and salinity properties, suggesting multiple source regions and enabling speculation of varying formation mechanisms.

The eddy field across the Arctic Ocean's Canada Basin is analyzed using Ice‐Tethered Profiler (ITP) and moored measurements of temperature, salinity, and velocity spanning 2005 to 2015. ITPs encountered 243 eddies, 98% of which were anticyclones, with approximately 70% of these having anomalously cold cores. The spatially and temporally varying eddy field is analyzed accounting for sampling biases in the unevenly distributed ITP data and caveats in detection methods. The highest concentration of eddies was found in the western and southern portions of the basin, close to topographic margins and boundaries of the Beaufort Gyre. The number of lower halocline eddies approximately doubled from 2005–2012 to 2013–2014. The increased eddy density suggests more active baroclinic instability of the Beaufort Gyre that releases available potential energy to balance the wind energy input; this may stabilize the Gyre spin‐up and associated freshwater increase.

A decade of moored measurements from the Arctic Ocean's northwestern Beaufort Gyre (collected as a component of the Beaufort Gyre Exploration Project) are analyzed to examine the range of mesoscale eddies over the water column and the dynamical processes that set eddy vertical scales. A total of 58 eddies were identified in the moored record, all anticyclones with azimuthal velocities ranging from 10 to 43 cm/s. These are divided into three classes based on core depths. Shallow eddies (core depths around 120 m) are shown to be vertically confined by the strong stratification of the halocline; typical thicknesses are around 100 m. Deep eddies (core depths around 1200 m) are much taller (thicknesses around 1300 m) owing to the weaker stratification at depth, consistent with a previous study. Eddies centered around mid‐depths all have two cores (vertically aligned and separated in depth) characterized by velocity maxima and anomalous temperature and salinity properties. One core is located at the base of the halocline (around 200 m depth) and the other at the depth of the Atlantic Water layer (around 400 m depth). These double‐core eddies have vertical scales between those of the shallow and deep eddies. The strongly decreasing stratification in their depth range motivates a derivation for the quasi‐geostrophic adjustment of a nonuniformly stratified water column to a potential vorticity anomaly. The result aids in interpreting the dynamics and origins of the double‐core eddies, providing insight into transport across a major water mass front separating Canadian and Eurasian Water.

Kinetic energy (KE) in the Arctic Ocean's Beaufort Gyre is dominated by the mesoscale eddy field that plays a central role in the transport of freshwater, heat, and biogeochemical tracers. Understanding Beaufort Gyre KE variability sheds light on how this freshwater reservoir responds to wind forcing and sea ice and ocean changes. The evolution and fate of mesoscale eddies relate to energy pathways in the ocean (e.g., the exchange of energy between barotropic and baroclinic modes). Mooring measurements of horizontal velocities in the Beaufort Gyre are analyzed to partition KE into barotropic and baroclinic modes and explore their evolution. We find that a significant fraction of water column KE is in the barotropic and the first two baroclinic modes. We explain this energy partitioning by quantifying the energy transfer coefficients between the vertical modes using the quasi-geostrophic potential vorticity conservation equations with a specific background stratification observed in the Beaufort Gyre. We find that the quasi-geostrophic vertical mode interactions uphold the persistence of KE in the first two baroclinic modes, consistent with observations. Our results explain the specific role of halocline structure on KE evolution in the gyre and suggest depressed transfer to the barotropic mode. This limits the capacity for frictional dissipation at the sea floor and suggests that energy dissipation via sea ice-ocean drag may be prominent.

The difference between North Atlantic subpolar gyre sea surface temperatures (SPG SSTs) and hemispheric-or global-scale surface temperatures has been utilized as an index of centennial-timescale changes in Atlantic Meridional Overturning Circulation (AMOC) strength. Here, using Community Earth System Model ensembles, we show that surface temperature-based indices (STIs) proposed to date largely reflect global-scale temperature trends and thus do not reflect dynamical relationships with AMOC. More broadly, we find that relationships between STIs, SPG SSTs, and AMOC strength differ greatly in significance and magnitude over different time periods because they are dependent upon the nature of external forcing. In the twentieth century, characterized by offsetting greenhouse gas and aerosol forcing, the relationship between SSTs and AMOC strength varies widely and changes sign across a 20-member ensemble. We conclude that STIs and SPG SSTs are poor predictors of centennial-timescale AMOC strength variations. Plain Language Summary The short observational record of the Atlantic Meridional Overturning Circulation (AMOC) limits our ability to assess changes in its strength over the instrumental and preinstrumental periods. Indirect proxies of ocean circulation are thus required to make inferences about past trends, for example, those over the past century. Several previous analyses have used surface temperature indices to interpret twentieth century AMOC trends. However, the robustness of this indirect AMOC proxy, including its sensitivity to time period, timescale, and/or climate state, has not been assessed. We use two state-of-the art climate model ensembles to assess AMOC/surface temperature relationships over century timescales, finding a strong dependence upon time period and climate forcing. Our results clarify the origins of discrepancies in AMOC/surface temperature relationships and suggest that interpretations of twentieth century climate and ocean circulation change based on surface temperature indices are limited. Over centennial timescales, many studies have noted a "warming hole" or "cold patch," in the North Atlantic subpolar gyre (SPG). In some model simulations, this SST pattern has been related to changes RESEARCH LETTER

No abstract

Ocean bottom pressure (p b ) variability reflects changes in mass of the water column arising from freshwater fluxes at the ocean boundaries and horizontal mass redistribution by the ocean circulation (e.g.,

Changes in dynamic manometric sea level ζm represent mass-related sea-level changes associated with ocean circulation and climate. We use twin model experiments to quantify magnitudes and spatiotemporal scales of ζm variability caused by barometric-pressure pa loading at long periods (≳ 1 month) and large scales (≳ 300 km) relevant to Gravity Recovery and Climate Experiment (GRACE) ocean data. Loading by pa drives basin-scale monthly ζm variability with magnitudes as large as a few cm. Largest ζm signals occur over abyssal plains, on the shelf, and in marginal seas. Correlation patterns of modeled ζm are determined by continental coasts and H/f contours (H is ocean depth and f is Coriolis parameter). On average, ζm signals forced by pa represent departures of ≲ 10% and ≲ 1% from the inverted-barometer effect ζib on monthly and annual periods, respectively. Basic magnitudes, spatial patterns, and spectral behaviors of ζm from the model are consistent with scaling arguments from barotropic potential vorticity conservation. We also compare ζm from the model driven by pa to ζm from GRACE observations. Modeled and observed ζm are significantly correlated across parts of the tropical and extratropical oceans, on shelf and slope regions, and in marginal seas. Ratios of modeled to observed ζm magnitudes are as large as ∼ 0.2 (largest in the Arctic Ocean) and qualitatively agree with analytical theory for the gain of the transfer function between ζm forced by pa and wind stress. Results demonstrate that pa loading is a secondary but nevertheless important contributor to monthly mass variability from GRACE over the ocean.