Horizontal sampling of the ocean has been sparse for decades because of technical limitations. This can contribute to an incomplete depiction and misleading understanding of the hydrography. This is a particular concern for complex submesoscale and smaller scale flow structures that influence stratification and vertical transport of properties.

We use high resolution observations from a Triaxus towed undulating vehicle and develop a statistical subsampling pipeline in order to present the first multi-scale investigation of subsurface and interior horizontal density variability in a global context. Hydrographic transects were performed between 2018 and 2022 with vertical ranges extending from near-surface values down to depths varying between 50 and 350m in the oceanographically distinct regimes of the Arctic marginal ice zone, of a coastal upwelling area, of the equatorial Atlantic, and of the Antarctic Circumpolar Current. The investigation of lateral density gradient fields follows a baseline spanning four orders of magnitude, from 2m to 25km. Our main objectives are to determine the scaling properties of density fronts and to identify oceanic regimes that are susceptible to an underestimation of their thermohaline variability.

We find that the amplitude of horizontal density gradients increases non-linearly as the horizontal resolution is increased, closely following a proposed power law over all observed scales. This relation is applicable throughout all study regions allowing for a potential prediction of the gradient distribution for scales not resolved by measurements. Submesoscale density gradients are of higher amplitude along the base of shallow mixed layers, and in the presence of subsurface currents, frontal systems, and eddies. The latter two create strong lateral anisotropies in the density field, masking other contributions to the multi-scale spread of gradients. Furthermore, the gradient fields are primarily driven by salinity variability at high northern latitudes and by temperature variability in regions closer to the equator; in the Southern Ocean temperature and salinity largely compensate. The decay rate of the estimated gradients with increasing horizontal distance is related to fractal properties and a scale-dependent compensation of the density field.

This highlights that there is a certain arbitrariness regarding the strengths of density gradients in the present literature. We recommend that the employed horizontal resolution always be quoted alongside values of the horizontal density gradient.

How to cite: Duong, B. L. and von Appen, W.-J.: Scale Dependence of Subsurface Horizontal Density Gradients as Observed In-Situ Across Four Orders of Magnitude, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-1518, https://doi.org/10.5194/egusphere-egu25-1518, 2025.

The Atlantic Meridional Overturning Circulation (AMOC) is a key component of the climate system, exhibiting strong variability across daily to millennial timescales and significantly influencing global climate. Sensitive to external conditions such as freshwater input, greenhouse gas concentrations, and aerosol forcing, important variations of the AMOC can be triggered by anthropogenic emissions. This study presents a comprehensive analysis of sources of AMOC variance in state-of-the-art climate ensemble models. By decomposing the effects of scenario, model, ensemble, and time variability, along with their interactions, through an Analysis of Variance (ANOVA), we identify three distinct regimes of AMOC variability from 1850 to 2100. The first regime, spanning most of the historical period, is characterized by a relatively stable AMOC dominated by internal variability. The second regime, initiated by AMOC decline at the end of the 20th century and lasting until mid-21st century, is governed by a transient increase of time variability. Notably, the direct effect of forcing differences remains muted all along this regime, despite the start of emission-scenarios in 2015. The third regime, beginning around 2050, is marked by the emergence and rapid dominance of inter-scenario variability. Throughout the simulations, model variability remains the primary source of uncertainty, influenced by aerosol forcing response, AMOC decline magnitude, and the physical variability. A key finding of this work is the evidence that internal variability decreases simultaneously with AMOC intensity and seems proportional to emission-scenario intensity. 

How to cite: Coquereau, A., Sévellec, F., Huck, T., Hirschi, J., and Jamet, Q.: Past, Present, and Future Variability of Atlantic Meridional Overturning Circulation in CMIP6 Ensembles, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-10891, https://doi.org/10.5194/egusphere-egu25-10891, 2025.

The Indonesian Throughflow (ITF) carries an annual average of about 15 Sv of water from the Pacific through the Indonesian Seas Into the Indian Ocean, and its year-to-year variation ranges from 1 to 4 Sv. A 10-member ensemble of 41-year integrations of a semi-global eddy-resolving oceanic general circulation model is examined to explore the intrinsic (chaotic) variability of the ITF transport and associated flow. It is found that the annual-mean ITF transport is different by about 1 Sv between the ensemble members at several years. The characteristic vertical and horizontal structures of the ensemble anomaly (deviation from the ensemble average) are described. These structures and the basin-scale spread of the anomaly suggest that the intrinsic variability of the ITF is a genuine increase or decrease of the classical ITF rather than variability due to local eddies or nonlinear currents within the Indonesian Seas. The lagged correlation of the intrinsic component of the ITF transport with sea-surface height and barotropic streamfunction suggests that the intrinsic variability may come from zonal jets in the western subtropical North Pacific.

How to cite: Furue, R., Nonaka, M., and Sasaki, H.: Intrinsic interannual variability of the Indonesian Throughflow, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-14639, https://doi.org/10.5194/egusphere-egu25-14639, 2025.

This study investigates the low-frequency variability of the Mediterranean Sea using an ensemble of 30 eddy-permitting (1/12°) NEMO-based regional ocean simulations. The ensemble members were slightly perturbed in their initial conditions and forced by the same atmospheric variability during 34 years, allowing us to separate the intrinsic and atmospherically-forced components of the ocean variability.

At interannual timescales, our analysis of sea surface height (SSH) reveals distinct patterns of intrinsic variability across the basin. While the variability of certain circulation features, such as the North Ionian Gyre (NIG), is mostly paced by the atmosphere, low-frequency fluctuations of other features —like in the Algerian Basin— are largely intrinsic and random. The variance decomposition reveals that intrinsic processes control most of the total SSH variability over one-fifth of the basin, highlighting their pivotal role in shaping the interannual fluctuations in the basin.

Inspired by previous studies of the Atlantic Meridional Overturning Circulation (Gregorio et al., 2015; Leroux et al., 2018), we investigate the forced and intrinsic components of the Mediterranean Zonal Overturning Circulation (ZOC) interannual variability, focusing on the eastward flow of Atlantic waters and westward flow of intermediate waters in density coordinates. While the transport in the western basin shows moderate variability, our results reveal an increase in total variability in the Levantine Basin, driven by both forced and intrinsic components. EOF analyses of ZOC fluctuations suggest distinct variability modes east and west of Sicily, which remain to be further investigated.

This work highlights the substantial contribution of intrinsic variability in various features of the Mediterranean's fluctuations, up to decadal timescales. A better understanding of the relative contributions of externally-driven and internally-generated oceanic fluctuations is crucial for accurately interpreting simulated and observed signals, making reliable predictions, and exploring possible impacts on marine ecosystems.

How to cite: Héron, D., Brankart, J.-M., and Brasseur, P.: Forced and Intrinsic Low-Frequency Variability in the Mediterranean Sea from a Multi-Decadal Ensemble Simulation, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20796, https://doi.org/10.5194/egusphere-egu25-20796, 2025.

In some places and over some time horizons, coastal sea-level is highly predictable. For example, there are locations where the seasonal cycle dominates and the monthly-mean sea- level is predictable for many years ahead. However, in other places, we know that there are frequent storm surges, that AMOC changes are linked to coastal sea-level, that mass anomalies propagate around the continental shelf slope boundary and can affect remote changes in coastal sea level, but also that there is a manifestation of internal or intrinsic, nonlinear processes which have a chaotic signature. From place to place, globally, there is a need to optimally predict coastal sea-level for societal planning and adaptation, to mitigate the effects of climate change and sea-level rise. However, on the regional and local scales, there are still many gaps, both in terms of observation and modelling of coastal sea-level on timescales relevant to people’s lives and wellbeing.

Using an ensemble modelling approach, one can use the ensemble mean and ensemble variance to estimate a potentially predictable,”forced” component of the system and a potentially unpredictable, ”unforced” component. In terms of sea-level, the unforced, chaotic intrinsic variability (CIV) component can, in some locations, dominate the forced component, even out to decadal timescales. This is known to be a major source of uncertainty in sea-level trends, relevant to IPCC projections, but analogously so for other temporal components on seasonal to decadal timescales too.

This study:

• a) verifies where and over which timescales of variability the OCCIPUT, eORCA025, 50- member initial condition ensemble simulation captures coastal sea-level from the GESLA3 tide gauge dataset.
• b) generates maps of the potential predictability of coastal sea-level.
• c) explores predictive suitability of statistical models versus GCMs.
• d) suggests relevant processes behind potential predictability characteristics.

How to cite: Wilson, C. and Williams, S. D. P.: Where, why, and over which timescales is coastal sea-level potentially predictable?, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-19468, https://doi.org/10.5194/egusphere-egu25-19468, 2025.

The Southwest Indian Ocean (SWIO) is characterized by diverse dynamic regimes, with intense energy fluxes and intricate atmospheric interactions (Phillips et al., 2021, OS). The Mascarene area, to the east of Madagascar, is influenced by the South Equatorial Current and the Indian subtropical gyre, the Mozambique Channel presents numerous mesoscale eddies, which play an important role in the biogeochemical dynamics, and the Equatorial zone is affected by the inversion of seasonal Monsoon circulation. Modeling such complex systems requires the consideration of multiple sources of uncertainty. In the context of global warming and climate projections, it is essential to simulate these uncertainties in order to obtain a more accurate representation and understanding of the SWIO ocean dynamics. The objective of this project is to identify and analyze the dominant sources of uncertainty affecting surface circulation in the SWIO. To address this issue, a probabilistic approach is integrated into the CROCO model (Coastal and Regional Ocean Community), following three key steps. Firstly, a realistic regional configuration of the CROCO model is developed for the SWIO region, which is forced and validated by CMEMS and ECMWF operational and satellite products. Then, a stochastic perturbation generator (referred to as STOGEN and originally developed in the NEMO model, Brankart et al., 2015, GMD) is implemented into CROCO, associated with an ensemble generator. Finally, several ensemble simulations are performed using stochastic processes with varying correlation structures in space and time within the defined regional setting. This allows to test the cumulative effect of different sources of uncertainty associated with surface ocean circulation by analyzing the ensemble statistics and variability based on surface variables such as sea surface height, temperature, salinity or velocity fields. We starts with the simulation of an ensemble by perturbing the wind stress. Then, three additional ensemble simulations will be generated by perturbing the vertical mixing, the initial conditions to analyze the intrinsic ocean variability and the open boundary conditions. The integration of stochastic parameterization within CROCO allow to simulate and partially quantify some of the non-deterministic effects of unresolved processes and scales. It enables an objective statistical comparison between model and observations associated with uncertainty description for data assimilation systems (Popov et al., 2024, OS).

How to cite: Weiss, L., Brankart, J.-M., Jamet, Q., and Brasseur, P.: A stochastic framework for modeling surface ocean variability in the Southwest Indian Ocean, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15591, https://doi.org/10.5194/egusphere-egu25-15591, 2025.

The non-linear nature of the ocean dynamics motivates the use of ensemble ocean simulations to discriminate the intrinsic and extrinsic sources of oceanic variability. Separating the mean and eddy flows from ensemble statistics also gives access to their local and instantaneous interactions in the non-stationary and inhomogeneous ocean. We here take advantage of this idea to quantify the local/instantaneous roles of laminar and eddy fluxes in the seasonal cycle of the North Atlantic subtropical mode water (STMW) that is formed through ocean-atmosphere interaction and controls large-scale oceanic ventilation.

We employ an ensemble of 48 North Atlantic 1/12-degree ocean simulations, where all members are driven by the same atmospheric forcing after slight initial perturbations. We achieve a space/time-dependent mean-eddy flow separation by obtaining a residual-mean flow that represents the common oceanic response of all ensemble members to the atmosphere, and a set of residual eddies that reflect the ensemble dispersion. We characterise the STMW as a low Ertel potential vorticity (PV) pool and find that its PV budget is mostly controlled by the ensemble mean PV flux: the formation and erosion of the STMW is predominantly driven by the residual-mean flow. The contribution of eddy PV transport is secondary; this can be attributed to the low intrinsic variability within the PV pool, as captured by the residual eddies. 

Overall, our results show that ensemble ocean simulations are powerful to investigate inhomogeneous, non-stationary, nonlinear multiscale ocean dynamics, providing deeper insights into the life cycle of large-scale climate-relevant features like STMW.

How to cite: Sun, L., Dewar, W., Deremble, B., Wienders, N., and Poje, A.: Ensemble Ocean Simulations in the North Atlantic: Exploring the Intrinsic Variability in Subtropical Mode Water Dynamics, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-12552, https://doi.org/10.5194/egusphere-egu25-12552, 2025.

When studying large tuburlent regions of the ocean, interactions between the mean flow and eddies plays a
central role in shaping large-scale circulation patterns by redistributing heat, momentum, and energy across the
ocean. Accurately representing these interactions in General Circulation Models (GCMs) remains a challenge,
particularly due to the subgrid-scale modelling issues and the limitations of traditional parameterization methods.
In this study we highlights the limitations of the diagnostics that can be performed with a too small in size en-
semble of simulations for capturing the Reynolds stress tensor as well as accurately diagnosing the work of such
tensor. In the context of studying energy exchange between the mean flow and eddies, the work of the Reynolds
stress is associated with the mean-to-eddy energy conversion rate MEC (Jamet et al. 2022).

To address the above issues, we explore the capabilities of the Location Uncertainty (LU) framework (Mémin
2014) to provide a better representation of eddy-mean flow energy transfer. By introducing stochastic variability
directly into the governing equations of fluid motion, LU provides an approach to model the unresolved turbulent
effects. A rederivation of the equation for the energy transfers is then possible through a stochastic version of
the Reynolds Transport Theorem (Bauer et al. 2020) and leads to an alternative representation of the interactions
between mean flow and eddies.

Based on 48-member ensemble simulation of the North Atlantic under realistic forcing, we provide a robust
comparison between deterministic and stochastic estimates of the work of Reynolds stress (MEC). By comparing
deterministic and stochastic estimates of MEC, we show that LU can effectively address the issues of stastistical
convergence by inflating intrinsic variability leading to a more robust representation of these non-linear terms. In
addition, statistical moments are shown to be more stable than from the deterministic formulation of the eddy-mean
flow interactions. Key results of this study include a detailed formulation of kinetic energy evolution equations
under the LU framework, which reveals significant improvements compared to the deterministic formulation of
the work of the Reynolds stress in terms of statistical moments. The noise definition relies, in this study, on the
snapshot proper orthogonal decomposition (POD) in the ensemble dimension, offering a time varying orthogonal
eigenfunctions basis. These diagnostics provide usefull tools to observe moving patterns and stability regions,
leading to physical interpretation of the eddy-mean flow interactions in the Gulf Stream.

How to cite: Nex, M., Jamet, Q., Mémin, E., and Sévellec, F.: A Stochastic description of eddy-mean flow interactions, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17449, https://doi.org/10.5194/egusphere-egu25-17449, 2025.

The ocean is forced by the fluxes of momentum, heat, and fresh water at the sea surface. When driving an ocean model using stationary fluxes for a sufficiently long time, we expect the model  to produce an equilibrated ocean characterized by stationary fluctuations. These fluctuations are not all synchronized with the surface fluxes.  For an ensemble obtained by forcing the ocean model with the same fluxes (starting from slightly different initial states),   fluctuations (at a time) that are  synchronized with the surface fluxes can be identified as the mean across the ensemble (at that time), and those not synchronized with the surface fluxes as the deviations from the mean across the ensemble. In case that the model has a sufficiently fine resolution, we expect that the latter — also known as intrinsic ocean variability — is substantial. The intrinsic ocean variability has to get its energy from somewhere. The only possible energy source is the surface fluxes, which originate from atmospheric motions supported (essentially) by the Sun. In this sense, intrinsic ocean variability can be considered as a feature of an ocean that is in equilibrium with a huge reservoir. 

The principle that governs equilibrium fluctuations — no matter how the equilibrium is reached — is a form of fluctuation-dissipation relation. The relation ensures that in an equilibrium with a reservoir, anything  that generates fluctuations must also dissipate fluctuations, and anything that dissipates fluctuations must also generate fluctuations. This principle makes a dynamical system in equilibrium with a reservoir be inherently random, even when the forcing resulting from the reservoir, such as the surface fluxes, is purely deterministic.  We evaluate this principle using solutions from the Lorenz's 1963 model and solutions obtained from the ICON ocean model with a horizontal resolution of  5 km. 

How to cite: von Storch, J.-S.: Principle of equilibrium fluctuations, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-7149, https://doi.org/10.5194/egusphere-egu25-7149, 2025.