Search for: All records

Recent years have seen a surge in interest for leveraging neural networks to parameterize small-scale or fast processes in climate and turbulence models. In this short paper, we point out two fundamental issues in this endeavor. The first concerns the difficulties neural networks may experience in capturing rare events due to limitations in how data is sampled. The second arises from the inherent multiscale nature of these systems. They combine high-frequency components (like inertia-gravity waves) with slower, evolving processes (geostrophic motion). This multiscale nature creates a significant hurdle for neural network closures. To illustrate these challenges, we focus on the atmospheric 1980 Lorenz model, a simplified version of the Primitive Equations that drive climate models. This model serves as a compelling example because it captures the essence of these difficulties.

 

Abstract Glacial fjord circulation modulates the connection between marine-terminating glaciers and the ocean currents offshore. These fjords exhibit a complex 3D circulation with overturning and horizontal recirculation components, which are both primarily driven by water mass transformation at the head of the fjord via subglacial discharge plumes and distributed meltwater plumes. However, little is known about the 3D circulation in realistic fjord geometries. In this study, we present high-resolution numerical simulations of three glacial fjords (Ilulissat, Sermilik, and Kangerdlugssuaq), which exhibit along-fjord overturning circulations similar to previous studies. However, one important new phenomenon that deviates from previous results is the emergence of multiple standing eddies in each of the simulated fjords, as a result of realistic fjord geometries. These standing eddies are long-lived, take months to spin up, and prefer locations over the widest regions of deep-water fjords, with some that periodically merge with other eddies. The residence time of Lagrangian particles within these eddies are significantly larger than waters outside of the eddies. These eddies are most significant for two reasons: 1) they account for a majority of the vorticity dissipation required to balance the vorticity generated by discharge and meltwater plume entrainment and act to spin down the overall recirculation and 2) if the eddies prefer locations near the ice face, their azimuthal velocities can significantly increase melt rates. Therefore, the existence of standing eddies is an important factor to consider in glacial fjord circulation and melt rates and should be taken into account in models and observations. 

Surface gravity wave effects on currents (WEC) cause the emergence of Langmuir cells (LCs) in a suite of high horizontal resolution (Δx= 30 m), realistic oceanic simulations in the open ocean of central California. During large wave events, LCs develop widely but inhomogeneously, with larger vertical velocities in a deeper mixed layer. They interact with extant submesoscale currents. A 550-m horizontal spatial filter separates the signals of LCs and of submesoscale and larger-scale currents. The LCs have a strong velocity variance with small density gradient variance, while submesoscale currents are large in both. Using coarse graining, we show that WEC induces a forward cascade of kinetic energy in the upper ocean up to at least a 5-km scale. This is due to strong positive vertical Reynolds stress (in both the Eulerian and the Stokes drift energy production terms) at all resolved scales in the WEC solutions, associated with large vertical velocities. The spatial filter elucidates the role of LCs in generating the shear production on the vertical scale of Stokes drift (10 m), while submesoscale currents affect both the horizontal and vertical energy fluxes throughout the mixed layer (50–80 m). There is a slightly weaker forward cascade associated with nonhydrostatic LCs (by 13% in average) than in the hydrostatic case, but overall the simulation differences are small. A vertical mixing schemeK-profile parameterization (KPP) partially augmented by Langmuir turbulence yields wider LCs, which can lead to lower surface velocity gradients compared to solutions using the standard KPP scheme.

 

Eddies play a crucial role in shaping ocean dynamics by affecting material transport, and generating spatio‐temporal heterogeneity. However, how eddies at different scales modulate biogeochemical transformation rates remains an open question. Applying a multi‐scale decomposition to a numerical simulation, we investigate the respective impact of mesoscale and submesoscale eddies on nutrient transport and biogeochemical cycling in the California Current System. First, the non‐linear nature of nutrient uptake by phytoplankton results in a 50% reduction in primary production in the presence of eddies. Second, eddies shape the vertical transport of nutrients with a strong compensation between mesoscale and submesoscale. Third, the eddy effect on nutrient uptake is controlled by the covariance of temperature, nutrient and phytoplankton fluctuations caused by eddies. Our results shed new light on the tight interaction between non‐linear fluid dynamics and ecosystem processes in realistic eddy regimes, which remain largely under‐resolved by global Earth system models.

 

Oceanic mixing, mostly driven by the breaking of internal waves at small scales in the ocean interior, is of major importance for ocean circulation and the ocean response to future climate scenarios. Understanding how internal waves transfer their energy to smaller scales from their generation to their dissipation is therefore an important step for improving the representation of ocean mixing in climate models. In this study, the processes leading to cross-scale energy fluxes in the internal wave field are quantified using an original decomposition approach in a realistic numerical simulation of the California Current. We quantify the relative contribution of eddy–internal wave interactions and wave–wave interactions to these fluxes and show that eddy–internal wave interactions are more efficient than wave–wave interactions in the formation of the internal wave continuum spectrum. Carrying out twin numerical simulations, where we successively activate or deactivate one of the main internal wave forcing, we also show that eddy–near-inertial internal wave interactions are more efficient in the cross-scale energy transfer than eddy–tidal internal wave interactions. This results in the dissipation being dominated by the near-inertial internal waves over tidal internal waves. A companion study focuses on the role of stimulated cascade on the energy and enstrophy fluxes.

 

Abstract A set of realistic coastal simulations in California allows for the exploration of surface gravity wave effects on currents (WEC) in an active submesoscale current regime. We use a new method that takes into account the full surface gravity wave spectrum and produces larger Stokes drift than the monochromatic peak-wave approximation. We investigate two high wave events lasting several days — one from a remotely generated swell and another associated with local wind-generated waves — and perform a systematic comparison between solutions with and without WEC at two submesoscale-resolving horizontal grid resolutions ( dx = 270 m and 100 m). WEC results in the enhancement of open-ocean surface density and velocity gradients when the averaged significant wave height H S is relatively large (> 4.2m). For smaller waves, WEC is a minor effect overall. For the remote swell (strong waves and weak winds), WEC maintains submesoscale structures and accentuates the cyclonic vorticity and horizontal convergence skewness of submesoscale fronts and filaments. The vertical enstrophy ζ 2 budget in cyclonic regions ( ζ/f > 2) reveals enhanced vertical shear and enstrophy production via vortex tilting and stretching. Wind-forced waves also enhance surface gradients, up to the point where they generate a small-submesoscale roll-cell pattern with high vorticity and divergence that extends vertically through the entire mixed layer. The emergence of these roll-cells results in a buoyancy gradient sink near the surface that causes a modest reduction in the typically large submesoscale density gradients.