Multi-timescale control of Southern Ocean diapycnal mixing over Atlantic tracer budgets

Oceanic cross-density (diapycnal) mixing helps sustain the ocean den- sity stratification and its Meridional Overturning Circulation (MOC) and is key to global tracer distributions. The Southern Ocean (SO) is a key region where different overturning cells connect, allowing nutri- ent and carbon rich Indian and Pacific deep waters, and oxygen rich Atlantic deep waters to resurface. The SO is also rife with localized intense diapycnal mixing due to breaking of internal waves induced by the interaction of energetic eddies and currents with rough topogra- phy. SO diapycnal mixing is believed to be of secondary importance for the MOC. Here we show that changes to SO mixing can cause sig- nificant alterations to Atlantic biogeochemical tracer distributions over short and long timescales in an idealized model of the MOC. While such alterations are dominated by the direct impact of changes in diapycnal mixing on tracer fluxes on annual to decadal timescales, on centennial timescales they are dominated by the mixing-induced variations in the advective transport of the tracers by the Atlantic MOC. This work sug- gests that an accurate representation of spatio-temporally variable local and non-local mixing processes in the SO is essential for climate mod- els’ ability to i) simulate the biogeochemical cycles and air sea carbon fluxes on decadal timescales, ii) represent the indirect impact of mixing- induced changes to MOC on biogeochemical cycles on longer timescales.

Parallel algorithms for solving diffusion-convection problems on a multiprocessor computer system using hybrid MPI / OpenMP technology

The paper deals with the mathematical models, algorithms and software for mathematical modeling of coastal systems’ water pollution spreading dynamics under various unfavorable phenomena of natural and artificial genesis, developed for high-performance cluster systems. Methods for partitioning the computational domain for solving diffusion-convection problems have been developed, which allow for efficient parallelization of a computationally complex modeling problem, taking into account the architecture of the multiprocessor system used. The developed mathematical models are based on high-precision models of hydrophysics and hydrobiology and take into account the peculiarities of water systems in the south of the Rostov region, as well as factors of hydrobiological dynamics such as microturbulent diffusion and advective transport in various directions, mechanisms of primary and secondary pollution of coastal systems, taking into account currents. The paper presents algorithms for solving a simulated problem based on MPI parallelization technology, as well as based on mixed MPI + OpenMP technology. Numerical experiments have been carried out and the two technologies efficiency comparison has been made in the conditions of computing cluster used.

Numerical Assessment of the Hybrid Approach for Simulating Three-Dimensional Flow and Advective Transport in Fractured Rocks

This study presents a hybrid approach for simulating flow and advective transport dynamics in fractured rocks. The developed hybrid domain (HD) model uses the two-dimensional (2D) triangular mesh for fractures and tetrahedral mesh for the three-dimensional (3D) rock matrix in a simulation domain and allows the system of equations to be solved simultaneously. This study also illustrates the HD model with two numerical cases that focus on the flow and advective transport between the fractures and rock matrix. The quantitative assessments are conducted by comparing the HD results with those obtained from the discrete fracture network (DFN) and equivalent continuum porous medium (ECPM) models. Results show that the HD model reproduces the head solutions obtained from the ECPM model in the simulation domain and heads from the DFN model in the fractures in the first case. The particle tracking results show that the mean particle velocity in the HD model can be 7.62 times higher than that obtained from the ECPM mode. In addition, the developed HD model enables detailed calculations of the fluxes at intersections between fractures and cylinder objects in the case and obtains relatively accurate flux along the intersections. The solutions are the key factors to evaluate the sources of contaminant released from the disposal facility.

Climate on the Blanca Massif, Sangre de Cristo Mountains, Colorado, USA, during the Last Glacial Maximum

Temperature-index modeling is used to determine the magnitude of temperature depression on the Blanca Massif, Colorado, required to maintain steady-state mass balances of nine reconstructed glaciers at their extent during the Last Glacial Maximum (LGM). The mean temperature depression thus determined is ~8.6 +0.7/−0.9 °C where the uncertainties account for those inherent in the glacier reconstructions, in model parameters (e.g., melt factors), and possible modest changes in LGM precipitation. Associated equilibrium-line altitudes (ELAs) exhibit a statistically significant directional dependency being lower toward the north and east. Under the assumption that regional temperature change was uniform, required changes in precipitation vary systematically—also exhibiting a directional dependency coinciding with that in ELAs—and indicate increases (over modern) occurred on the eastern side of the massif while decreases occurred on the western side. This disparity represents a strengthening of a precipitation asymmetry, particularly winter precipitation, which exists today. The modern precipitation asymmetry may be a consequence of snow being blown over to the eastern side of the massif (advective transport) by southwesterly flow. Intensification of this flow during the LGM would have enhanced advection, and augmented snow accumulation on glaciers, thus explaining the lower ELAs and increased precipitation on that side of the massif.

Macroscopic Model for Passive Mass Dispersion in Porous Media Including Knudsen and Diffusive Slip Effects

In this work, a macroscopic model for incompressible and Newtonian gas flow coupled to Fickian and advective transport of a passive solute in rigid and homogeneous porous media is derived. At the pore-scale, both momentum and mass transport phenomena are coupled, not only by the convective mechanism in the mass transport equation, but also in the solid-fluid interfacial boundary condition. This boundary condition is a generalization of the Kramers-Kistemaker slip condition that includes the Knudsen effects. The resulting upscaled model, applicable in the bulk of the porous medium, corresponds to: 1) A Darcy-type model that involves an apparent permeability tensor, complemented by a dispersive term and 2) A macroscopic convection-dispersion equation for the solute, in which both the macroscopic velocity and the total dispersion tensor are influenced by the slip effects taking place at the pore-scale. The use of the model is restricted by the starting assumptions imposed in the governing equations at the pore scale and by the (spatial and temporal) constraints involved in the upscaling process. The different regimes of application of the model, in terms of the Péclet number values, are discussed as well as its extents and limitations. This new model generalizes previous attempts that only include either Knudsen or diffusive slip effects in porous media.

Advection Not Dispersion and Transient Storage Controls Streambed Nutrient and Greenhouse Gas Concentrations

Streams and rivers are globally important in the carbon and nitrogen cycles due to high carbon and nitrogen turnover rates and contribute disproportionately to global greenhouse gas (GHG) emissions relative to their areal coverage. The hyporheic zone may be a hotspot of biogeochemical reactivity within fluvial ecosystems resulting in high rates of nutrient attenuation and associated GHG production. Controls on streambed nutrient cycling and particularly GHG production remain insufficiently understood. In this study, porewater concentrations of nutrients (NH4+, NO3-, NO2-) and GHGes (CO2, CH4, N2O) were measured alongside surface water breakthrough curves (BTC) of conservative (uranine) and reactive tracers [resazurin (raz)-resorufin (rru)] to provide insights into often assumed correlations between in-stream advective transport and transient storage metrics, and streambed biogeochemistry. Streambed biogeochemical concentrations were significantly correlated with advective transport time but not with dispersion and transient storage. The effect of advective transport time varied between chemical species, with NH4+, CO2, and CH4 exhibiting positive correlations and NO3-, NO2-, and N2O displaying negative correlations with advective transport time and vice versa for long-term storage. These findings increase knowledge of the relationship between hydrological drivers and streambed chemistry, potentially highlighting areas of the streambed expected to have elevated nutrients and GHGs. This improved understanding may allow chemical species to be effectively targeted by morphological restoration, which will aid in effective pollution and climate remediation.

Deep learning predicts boiling heat transfer

Boiling is arguably Nature’s most effective thermal management mechanism that cools submersed matter through bubble-induced advective transport. Central to the boiling process is the development of bubbles. Connecting boiling physics with bubble dynamics is an important, yet daunting challenge because of the intrinsically complex and high dimensional of bubble dynamics. Here, we introduce a data-driven learning framework that correlates high-quality imaging on dynamic bubbles with associated boiling curves. The framework leverages cutting-edge deep learning models including convolutional neural networks and object detection algorithms to automatically extract both hierarchical and physics-based features. By training on these features, our model learns physical boiling laws that statistically describe the manner in which bubbles nucleate, coalesce, and depart under boiling conditions, enabling in situ boiling curve prediction with a mean error of 6%. Our framework offers an automated, learning-based, alternative to conventional boiling heat transfer metrology.