Mesoscale anticyclonic eddies dominate the sea-surface height variability in the Caribbean Sea. Although it is well established that these anticyclones are formed near the eastern boundary of the Caribbean Sea, which is demarcated by the Lesser Antilles, the source of their anticyclonic vorticity remains unclear. To gain insight into this source, we analyze the fluxes of vorticity into the Caribbean at its eastern boundary using a high-resolution numerical model. We find that the anticyclonic vorticity in the eastern Caribbean Sea predominantly originates from regions where intense ocean currents flow close to the Lesser Antilles. More specifically, St. Lucia and Grenada are hotspots for vorticity generation. The local generation rate scales with the amplitude of the volume transport through the passages between these islands. This finding is in contrast with the view that anticyclonic North Brazil Current (NBC) rings in the Atlantic Ocean are the main source of anticyclonic vorticity in the eastern Caribbean Sea. Our analyses reveal that the direct contribution of the vorticity of the NBC rings is of lesser importance than the local generation. However, the collision of upstream NBC rings with the Lesser Antilles increases the volume transport through the passages into the Caribbean Sea, so that their presence indirectly leads to enhanced local production of anticyclonic vorticity. This process is an example of the importance of vorticity generation near topography, which is ubiquitous in the oceans, and expected to be important whenever currents and steep topography meet.@en

Railway systems occasionally get into a state of being out-of-control, meaning that barely any train is running, even though the required resources (infrastructure, rolling stock and crew) are available. Because of the large number of affected resources and the absence of detailed, timely and accurate information, currently existing disruption management techniques cannot be applied in out-of-control situations. Most of the contemporary approaches assume that there is only one single disruption with a known duration, that all information about the resources is available, and that all stakeholders in the operations act as expected. Another limitation is the lack of knowledge about why and how disruptions accumulate and whether this process can be predicted. To tackle these problems, we develop a multidisciplinary framework combining techniques from complexity science and operations research, aiming at reducing the impact of these situations and—if possible—avoiding them. The key elements of this framework are (i) the generation of early warning signals for out-of-control situations, (ii) isolating a specific region such that delay stops propagating, and (iii) the application of decentralized decision making, more suited for information-sparse out-of-control situations.@en

Here, we describe our efforts to create a multi-scale and multi-physics framework that can be retargeted across different disciplines. Currently we have implemented our approach in the astrophysical domain, for which we developed AMUSE (github.com/amusecode/amuse ), and generalized this to the oceanographic and climate sciences, which led to the development of OMUSE (bitbucket.org/omuse ). The objective of this paper is to document the design choices that led to the successful implementation of these frameworks as well as the future challenges in applying this approach to other domains.@en

Here, we describe our efforts to create a multi-scale and multi-physics framework that can be retargeted across different disciplines. Currently we have implemented our approach in the astrophysical domain, for which we developed AMUSE (github.com/amusecode/amuse ), and generalized this to the oceanographic and climate sciences, which led to the development of OMUSE (bitbucket.org/omuse ). The objective of this paper is to document the design choices that led to the successful implementation of these frameworks as well as the future challenges in applying this approach to other domains.@en

The Yucatan Channel connects the Caribbean Sea with the Gulf of Mexico and is the main outflow region of the Caribbean Sea. Moorings in the Yucatan Channel show high-frequent variability in kinetic energy (50–100 days) and transport (20–40 days), but the physical mechanisms controlling this variability are poorly understood. In this study, we show that the short-term variability in the Yucatan Channel transport has an upstream origin and arises from processes in the North Brazil Current. To establish this connection, we use data from altimetry and model output from several high resolution global models. A significant 40–70 day variability is found in the sea surface height in the North Brazil Current retroflection region with a propagation toward the Lesser Antilles. The frequency of variability is generated by intrinsic processes associated with the shedding of eddies, rather than by atmospheric forcing. This sea surface height variability is able to pass the Lesser Antilles, it propagates westward with the background ocean flow in the Caribbean Sea and finally affects the variability in the Yucatan Channel volume transport.@en

Because of its pronounced fresh signature, the properties of the northward-flowing Antarctic Intermediate Water (AAIW) affect the Atlantic Meridional Overturning Circulation. Hence, understanding modifications of AAIW along its path is important. Here, we analyze AAIW changes along its path in the Caribbean Sea and assess whether vertical fluxes from background turbulence and from double-diffusive mixing in thermohaline staircases can explain these variations. We deduce the occurrence rate of staircases (7%) and estimate the flux ratio (Formula presented.) from Argo float profiles. In combination with vertical fluxes from background turbulence, these values are used in a steady-state advection-diffusion model to estimate the effective diffusivity of salt that arises from double diffusion (Formula presented.). This value for (Formula presented.) is similar to observed values (Schmitt, 2005, https://doi.org/10.1126/science.1108678), implying the observed modification of AAIW in the Caribbean Sea may be attributable primarily to vertical mixing in the region itself.@en

Railway systems occasionally get into a state of out-of-control, meaning that there is barely any train is running, even though the required resources (infrastructure, rolling stock and crew) are available. These situations can either be caused by large disruptions or unexpected propagation and accumulation of delays. Because of the large number of affected resources and the absence of detailed, timely and accurate information, currently existing methods cannot be applied in out-of-control situations. Most of the contemporary approaches assume that there is only one single disruption with a known duration, that all information about the resources is available, and that all stakeholders in the operations act as expected. Another limitation is the lack of knowledge about why and how disruptions accumulate and whether this process can be predicted. To tackle these problems, we develop a multidisciplinary framework aiming at reducing the impact of these situations and - if possible - avoiding them. The key elements of this framework are (i) the generation of early warning signals for out-of-control situations using tools from complexity science and (ii) a set of rescheduling measures robust against the features of out-of-control situations, using tools from operations research. @en

In present-day forward time stepping ocean-climate models, capturing both the wind-driven and thermohaline components, a substantial amount of CPU time is needed in a so-called spin-up simulation to determine an equilibrium solution. In this paper, we present methodology based on Jacobian-Free Newton-Krylov methods to reduce the computational time for such a spin-up problem. We apply the method to an idealized configuration of a state-of-the-art ocean model, the Modular Ocean Model version 4 (MOM4). It is shown that a typical speed-up of a factor 10-25 with respect to the original MOM4 code can be achieved and that this speed-up increases with increasing horizontal resolution.@en

To understand the physics and dynamics of the ocean circulation, techniques of numerical bifurcation theory such as continuation methods have proved to be useful. Up to now these techniques have been applied to models with relatively few (O (105)) degrees of freedom such as multi-layer quasi-geostrophic and shallow-water models and relatively low-resolution (e.g., 4° horizontal resolution) primitive equation models. In this paper, we present a new approach in which continuation methods are combined with parallel numerical linear system solvers. With this implementation, we show that it is possible to compute steady states versus parameters (and perform fully implicit time integration) of primitive equation ocean models with up to a few million degrees of freedom.@en

Thermohaline staircases are associated with double-diffusive mixing. They are characterized by stepped structures consisting of mixed layers of typically tens of metres thick that are separated by much thinner interfaces. Through these interfaces enhanced diapycnal salt and heat transport take place. In this study, we present a global dataset of thermohaline staircases derived from observations of Argo profiling floats and Ice-Tethered Profilers using a novel detection algorithm. To establish the presence of thermohaline staircases, the algorithm detects subsurface mixed layers and analyses the interfaces in between. Of each detected staircase, the conservative temperature, absolute salinity, depth, and height, as well as some other properties of the mixed layers and interfaces, are computed. The algorithm is applied to 487 493 quality-controlled temperature and salinity profiles to obtain a global dataset. The performance of the algorithm is verified through an analysis of independent regional observations. The algorithm and global dataset are available at https://doi.org/10.5281/zenodo.4286170. @en

Joint effects of the dynamic sea-level rise projected changes in the large-scale atmosphere/ocean circulation, and wave climate on hurricane-induced extreme water levels in the Caribbean region are assessed. We use the 2D-depth integrated ADCIRC + SWAN wave-ocean model, baroclinically coupled to an ocean-eddying version of the Community Earth System Model, to compare impacts of the September 2017 hurricanes with projected impacts of similar hypothetical tropical storms occurring in the future. The model predicts only minor changes in the hurricane-induced extreme water levels for those Caribbean islands which were severely devastated by the 2017 tropical storms (Irma and Maria). That is, provided that the hurricane intensity remains at the present-day level, the global mean sea-level rise is the main future coastal flood risk factor.@en

In this paper we present the Oceanographic Multipurpose Software Environment (OMUSE). OMUSE aims to provide a homogeneous environment for existing or newly developed numerical ocean simulation codes, simplifying their use and deployment. In this way, numerical experiments that combine ocean models representing different physics or spanning different ranges of physical scales can be easily designed. Rapid development of simulation models is made possible through the creation of simple high-level scripts. The low-level core of the abstraction in OMUSE is designed to deploy these simulations efficiently on heterogeneous high-performance computing resources. Cross-verification of simulation models with different codes and numerical methods is facilitated by the unified interface that OMUSE provides. Reproducibility in numerical experiments is fostered by allowing complex numerical experiments to be expressed in portable scripts that conform to a common OMUSE interface. Here, we present the design of OMUSE as well as the modules and model components currently included, which range from a simple conceptual quasi-geostrophic solver to the global circulation model POP (Parallel Ocean Program). The uniform access to the codes' simulation state and the extensive automation of data transfer and conversion operations aids the implementation of model couplings. We discuss the types of couplings that can be implemented using OMUSE. We also present example applications that demonstrate the straightforward model initialization and the concurrent use of data analysis tools on a running model. We give examples of multiscale and multiphysics simulations by embedding a regional ocean model into a global ocean model and by coupling a surface wave propagation model with a coastal circulation model.@en

The mesoscale variability in the Caribbean Sea is dominated by anticyclonic eddies that are formed in the eastern part of the basin. These anticyclones intensify on their path westward while they pass the coastal upwelling region along the Venezuelan and Colombian coast. In this study, we used a regional model to show that this westward intensification of Caribbean anticyclones is steered by the advection of cold upwelling filaments. Following the thermal wind balance, the increased horizontal density gradients result in an increase in the vertical shear of the anticyclones and in their westward intensification. To assess the impact of variations in upwelling on the anticyclones, several simulations were performed in which the northward Ekman transport (and thus the upwelling strength) is altered. As expected, stronger (weaker) upwelling is associated with stronger (weaker) offshore cooling and a stronger (weaker) westward intensification of the anticyclones. Moreover, the simulations with weaker upwelling show farther advection of the Amazon and Orinoco River plumes into the basin. As a result, in these simulations the horizontal density gradients were predominantly set by horizontal salinity gradients. The importance of the horizontal density gradients driven by temperature, which are associated with the upwelling, increased with increasing upwelling strength. The results of this study highlight that both upwelling and the advection of the river plumes affect the life cycle of mesoscale eddies in the Caribbean Sea.@en

Studies on the palaeoclimate and palaeoceanography using numerical model simulations may be considerably dependent on the implemented geographical reconstruction. Because building the palaeogeographic datasets for these models is often a time-consuming and elaborate exercise, palaeoclimate models frequently use reconstructions in which the latest state-of-the-art plate tectonic reconstructions, palaeotopography and -bathymetry, or vegetation have not yet been incorporated. In this paper, we therefore provide a new method to efficiently generate a global geographical reconstruction for the middle-late Eocene. The generalised procedure is also reusable to create reconstructions for other time slices within the Cenozoic, suitable for palaeoclimate modelling. We use a plate-tectonic model to make global masks containing the distribution of land, continental shelves, shallow basins and deep ocean. The use of depth-age relationships for oceanic crust together with adjusted present-day topography gives a first estimate of the global geography at a chosen time frame. This estimate subsequently needs manual editing of areas where existing geological data indicate that the altimetry has changed significantly over time. Certain generic changes (e.g. lowering mountain ranges) can be made relatively easily by defining a set of masks while other features may require a more specific treatment. Since the discussion regarding many of these regions is still ongoing, it is crucial to make it easy for changes to be incorporated without having to redo the entire procedure. In this manner, a complete reconstruction can be made that suffices as a boundary condition for numerical models with a limited effort. This facilitates the interaction between experts in geology and palaeoclimate modelling, keeping reconstructions up to date and improving the consistency between different studies. Moreover, it facilitates model inter-comparison studies and sensitivity tests regarding certain geographical features as newly generated boundary conditions can more easily be incorporated in different model simulations. The workflow is presented covering a middle-late Eocene reconstruction (38ĝ€Ma), using a MatLab script and a complete set of source files that are provided in the supplementary material.@en

Previous studies have indicated that most of the net sinking associated with the downward branch of the Atlantic Meridional Overturning Circulation (AMOC) must occur near the subpolar North Atlantic boundaries. In this work we have used monthly mean fields of a high-resolution ocean model (0.1 at the Equator) to quantify this sinking. To this end we have calculated the Eulerian net vertical transport (WΣ) from the modeled vertical velocities, its seasonal variability, and its spatial distribution under repeated climatological atmospheric forcing conditions. Based on this simulation, we find that for the whole subpolar North Atlantic WΣ peaks at about -14 Sv at a depth of 1139 m, matching both the mean depth and the magnitude of the meridional transport of the AMOC at 45° N. It displays a seasonal variability of around 10 Sv. Three sinking regimes are identified according to the characteristics of the accumulated WΣ with respect to the distance to the shelf: one within the first 90 km and onto the bathymetric slope at around the peak of the boundary current speed (regime I), the second between 90 and 250 km covering the remainder of the shelf where mesoscale eddies exchange properties (momentum, heat, mass) between the interior and the boundary (regime II), and the third at larger distances from the shelf where is mostly driven by the ocean's interior eddies (regime III). Regimes I and II accumulate 90% of the of the total sinking and display smaller seasonal changes and spatial variability than regime III. We find that such a distinction in regimes is also useful to describe the characteristics of WΣ in marginal seas located far from the overflow areas, although the regime boundaries can shift a few tens of kilometers inshore or offshore depending on the bathymetric slope and shelf width of each marginal sea. The largest contributions to the sinking come from the Labrador Sea, the Newfoundland region, and the overflow regions. The magnitude, seasonal variability, and depth at which peaks vary for each region, thus revealing a complex picture of sinking in the subpolar North Atlantic.@en

Sea-level rise poses severe threats to coastal and low-lying regions around the world, by exacerbating coastal erosion and flooding. Adequate sea-level projections over the next decades are important for both decision making and for the development of successful adaptation strategies in these coastal and low-lying regions to climate change. Ocean components of climate models used in the most recent sea-level projections do not explicitly resolve ocean mesoscale processes. Only a few effects of these mesoscale processes are represented in these models, which leads to errors in the simulated properties of the ocean circulation that affect sea-level projections. Using the Caribbean Sea as an example region, we demonstrate a strong dependence of future sea-level change on ocean model resolution in simulations with a global climate model. The results indicate that, at least for the Caribbean Sea, adequate regional projections of sea-level change can only be obtained with ocean models which capture mesoscale processes.@en

We introduce the Oceanographic Multipurpose Software Environment ( OMUSE): an open source framework for oceanographic simulation codes developed at the IMAU (Utrecht) using coupling technology developed at Leiden Observatory (Leiden). OMUSE aims to provide a homogeneous environment for numerical ocean simulation codes, simplifying their use and deployment. Using OMUSE numerical experiments that combine ocean models representing different physics or spanning different ranges of physical scales can be easily designed. Here, we present the design of OMUSE as well as the modules and model components currently included, which range from a simple conceptual models to to full global circulation models such as POP. We discuss the types of the couplings that can be implemented using OMUSE and present examples of OMUSE applications, that demonstrate the efficient and relatively straightforward model initialisation and coupling possible with OMUSE.@en

The Atlantic Meridional Overturning Circulation (AMOC) plays an important role in climate. The classical view of an ocean conveyor belt with northward surface currents and southward return currents transporting convectively-formed waters from the subpolar North Atlantic Ocean to other ocean basins suggests a tight relation between convection and sinking. However, convection regions feature very little vertical mass transport. Instead, it has been argued that the sinking of waters must take place near boundaries where ageostrophic processes affect the flow. So far, this has been confirmed in highly idealized regional model studies and in laboratory experiments. It is, however, unclear how well the sinking of dense waters is represented in the current generation of global ocean models and climate models, and whether the factors driving and controlling the sinking in these models are in accordance with the developed theory. This is of crucial importance for our confidence in projections of the future behavior of the AMOC, which are based on this type of model. In this study, we analyze the characteristics of the vertical transport in two global ocean models: an eddy-permitting model at 0.25 degree resolution and its coarser 1.0 degree resolution counterpart. We show that the sinking indeed predominantly occurs in a narrow region close to the boundary in both model simulations, and not in deep convection regions. Notably, the amount of vertical transport that is found along the edges of the North Atlantic Ocean is highly variable in space, and large differences exist between the two model versions. In the eddy-permitting model, the magnitude of the local sinking appears to be governed by alongshore variations in density near the boundary, in line with theory@en