How does stably-stratified shear-driven turbulence mix our oceans and estuaries?
Lead Research Organisation:
University of Cambridge
Department Name: Applied Maths and Theoretical Physics
Abstract
This research is ultimately motivated by reducing the harmful consequences of climate change on society, in the UK and worldwide.
The root of the problem is global warming, caused by the greenhouse effect of carbon dioxide from fossil fuels. As our atmosphere warms, so do our oceans, which directly affects biodiversity and causes sea levels to rise. As our oceans warm, the balance of forces that keep them in constant motion changes too, disrupting their worldwide circulation. This disruption is worrying, both in the short and long term, because the present circulation patterns perform at least two functions vital to our hospitable climate. First, vertical currents store excess heat and carbon deep into the ocean (slowing global warming). Second, North-South currents redistribute tropical heat to more temperate regions (reducing extreme weather and climate). Therefore, a weakening of these currents could accelerate climate change, with long-lasting societal consequences.
To mitigate this, scientists try to predict how the world's climate will evolve by using advanced mathematical and computer models of the ocean circulation. However, these models and their predictions need to be improved to be of greater benefit to society and decision-makers. A serious cause of uncertainty in these models lies in the mixing between water currents that have different salinity or temperature (and thus density). Currents of different densities organise into vertically-stacked (or "stably-stratified") layers which flow past one another at different speeds (creating a "shear" flow). These flows are always turbulent, which means that a vast number of tiny chaotic "eddies" mix the salinity and temperature of much larger layers in complex and unpredictable ways.
This fundamental but extremely challenging process of turbulent mixing in stably-stratified shear flows needs to be better understood. To do this, I will employ the following scientific approach in three steps.
First, I will use an accurate, reduced-scale model of such flows in the laboratory. This has two great benefits: it gives full control over the flow geometry, the density difference, flow speed, etc, allowing me to test and understand the influence of each parameter separately; and it allows me to use high-tech measurements to quantify the chaotic eddies and their mixing better than ever before.
Second, I will interpret these new laboratory measurements with mathematical models of turbulent mixing to generalise (or "extrapolate") my findings to real-scale flows found in the ocean. This crucial step relies on the power of "dimensional analysis" in fluid dynamics, which is also routinely used by engineers to develop new aircraft or ship designs from smaller-scale laboratory prototypes.
Third, I will verify the validity of my real-scale predictions by comparing them to actual measurements taken from ships (which are usually sparse, expensive, and less accurate). This step is similar to engineers performing a full-scale test before production, except that we have no control over the ocean. Although challenging, this "validation" step will help ensure that my whole approach succeeds in providing climate scientists with more accurate models for ocean mixing.
In addition to the long-term effects of global warming, I will also apply the above three steps to a shorter-term consequence: saltwater intrusions in estuaries. Sea level rise, more frequent droughts, extreme storm surges, and stronger tides will all increase the gradual encroachment of seawater in densely-populated deltas (including the important Thames Basin in the UK). The upstream intrusion of a dense saltwater layer beneath the fresh river water, and their vertical mixing reduce the availability of surface freshwater, with devastating consequences for coastal communities already felt around the world. I will develop more accurate models of mixing in saltwater intrusions to help mitigate this urgent problem.
The root of the problem is global warming, caused by the greenhouse effect of carbon dioxide from fossil fuels. As our atmosphere warms, so do our oceans, which directly affects biodiversity and causes sea levels to rise. As our oceans warm, the balance of forces that keep them in constant motion changes too, disrupting their worldwide circulation. This disruption is worrying, both in the short and long term, because the present circulation patterns perform at least two functions vital to our hospitable climate. First, vertical currents store excess heat and carbon deep into the ocean (slowing global warming). Second, North-South currents redistribute tropical heat to more temperate regions (reducing extreme weather and climate). Therefore, a weakening of these currents could accelerate climate change, with long-lasting societal consequences.
To mitigate this, scientists try to predict how the world's climate will evolve by using advanced mathematical and computer models of the ocean circulation. However, these models and their predictions need to be improved to be of greater benefit to society and decision-makers. A serious cause of uncertainty in these models lies in the mixing between water currents that have different salinity or temperature (and thus density). Currents of different densities organise into vertically-stacked (or "stably-stratified") layers which flow past one another at different speeds (creating a "shear" flow). These flows are always turbulent, which means that a vast number of tiny chaotic "eddies" mix the salinity and temperature of much larger layers in complex and unpredictable ways.
This fundamental but extremely challenging process of turbulent mixing in stably-stratified shear flows needs to be better understood. To do this, I will employ the following scientific approach in three steps.
First, I will use an accurate, reduced-scale model of such flows in the laboratory. This has two great benefits: it gives full control over the flow geometry, the density difference, flow speed, etc, allowing me to test and understand the influence of each parameter separately; and it allows me to use high-tech measurements to quantify the chaotic eddies and their mixing better than ever before.
Second, I will interpret these new laboratory measurements with mathematical models of turbulent mixing to generalise (or "extrapolate") my findings to real-scale flows found in the ocean. This crucial step relies on the power of "dimensional analysis" in fluid dynamics, which is also routinely used by engineers to develop new aircraft or ship designs from smaller-scale laboratory prototypes.
Third, I will verify the validity of my real-scale predictions by comparing them to actual measurements taken from ships (which are usually sparse, expensive, and less accurate). This step is similar to engineers performing a full-scale test before production, except that we have no control over the ocean. Although challenging, this "validation" step will help ensure that my whole approach succeeds in providing climate scientists with more accurate models for ocean mixing.
In addition to the long-term effects of global warming, I will also apply the above three steps to a shorter-term consequence: saltwater intrusions in estuaries. Sea level rise, more frequent droughts, extreme storm surges, and stronger tides will all increase the gradual encroachment of seawater in densely-populated deltas (including the important Thames Basin in the UK). The upstream intrusion of a dense saltwater layer beneath the fresh river water, and their vertical mixing reduce the availability of surface freshwater, with devastating consequences for coastal communities already felt around the world. I will develop more accurate models of mixing in saltwater intrusions to help mitigate this urgent problem.
Organisations
- University of Cambridge (Lead Research Organisation)
- University of Chicago (Collaboration)
- University of Massachusetts Amherst (Collaboration)
- York University Toronto (Collaboration)
- Woods Hole Oceanographic Institution (Collaboration)
- Arizona State University (Collaboration)
- Eindhoven University of Technology (Collaboration)
Publications
Atoufi A
(2023)
Stratified inclined duct: two-layer hydraulics and instabilities
in Journal of Fluid Mechanics
Atoufi A
(2023)
Stratified inclined duct: Two-layer hydraulics and instabilities
Jiang X
(2023)
Geometry of stratified turbulent mixing: local alignment of the density gradient with rotation, shear and viscous dissipation
in Journal of Fluid Mechanics
Lefauve A
(2025)
Routes to stratified turbulence and temporal intermittency revealed by a cluster-based network model of experimental data
in Europhysics Letters
Lefauve A
(2024)
Data-driven classification of sheared stratified turbulence from experimental shadowgraphs
in Physical Review Fluids
Lefauve A
(2024)
Geophysical stratified turbulence and mixing in the laboratory
in Comptes Rendus. Physique
Zhu L
(2023)
Stratified inclined duct: direct numerical simulations
in Journal of Fluid Mechanics
Related Projects
| Project Reference | Relationship | Related To | Start | End | Award Value |
|---|---|---|---|---|---|
| NE/W008971/1 | 31/03/2023 | 28/02/2025 | £702,235 | ||
| NE/W008971/2 | Transfer | NE/W008971/1 | 01/03/2025 | 30/03/2028 | £432,380 |
| Title | Research data supporting "Data-driven classification of sheared stratified turbulence from experimental shadowgraphs" |
| Description | For a more detailed description, refer to READ_ME.pdf This dataset corresponds to an automated analysis of stratified turbulent flow visualisations in an inclined duct laboratory experiment. The goal is to reduce the complexity of 113 long shadowgraph movies previously shared on another repository doi.org/10.17863/CAM.104471. Our analysis takes each of the 50,155 frames and automatically extracts statistics corresponding to the morphology of density interfaces embeddd within the flow, before clustering these statistics in a low-dimensional space. The resulting clusters are physically interpretable and allow us to classify the flow instantaneously into a distinct type of turbulence. The main dataset is the Matlab array pca_and_clustering_data.mat, providing the interface morphology 10-D vector and the ensuing clustering of each of the 50,155 frames processed in the associated article (see the methodology and variable names in Section III of the article). In addition, we provide: - the archive movies.zip contains six .mp4 movies showing the temporal evolution of all experiments at theta=1, 2, 3, 4, 5 and 6 degrees (including the principal components and clusters). Note that clusters initially named A, B, C, D, E in the movies correspond to L, B, O, G, U respectively in the paper. - metadata_spreadsheet.xlsx shows all the experimental parameters and information about temporal spacing between frames. - clustering_code.zip contains the Matlab code used to cluster the data (including the implementation of the OPTICS algorithm). |
| Type Of Material | Database/Collection of data |
| Year Produced | 2024 |
| Provided To Others? | Yes |
| Impact | Dataset was downloaded 29 times only a month after being live. |
| URL | https://www.repository.cam.ac.uk/handle/1810/364548 |
| Title | Research data supporting "Stratified inclined duct: direct numerical simulations" |
| Description | Three-dimensional DNS data and associated movies (gifs) for the five main cases B2, B5, B6, B8 and B10 in the paper "Stratified inclined duct: direct numerical simulations". The reservoirs have been cropped out so that only the duct domain is kept, i.e. (x,y,z) \in [-30,30]x[-1,1]x[-1,1]. Each dataset B2, B5, B6, B8, B10 consists of 46 snapshots starting at t=80 (after any initial transients) and until 260, spaced by 4 (nondimensional advective time units). B2-B8 are provided at the resolution 92565121 (as in Table 1 of the paper, noting only the duct is kept, without reservoirs). B10 is provided at 128565145 (given the higher Re=1000). Additionally we provide the high-resolution version of B10 at 2571121121 (as in Table 1), but at a reduced temporal resolution (16 snapshots from t=80 to 260 spaced by 12). Please contact the authors directly if you wish to obtain data at a higher temporal resolution or other data discussed in the paper. |
| Type Of Material | Database/Collection of data |
| Year Produced | 2023 |
| Provided To Others? | Yes |
| Impact | Viewed 59 times since deposit. |
| URL | https://www.repository.cam.ac.uk/handle/1810/353731 |
| Title | Shadowgraph visualisations of salt-stratified turbulence obtained in a stratified inclined duct (SID) laboratory experiment |
| Description | A collection of 113 experimental datasets corresponding to flow visualisations obtained in the G. K. Batchelor Laboratory (Department of Applied Mathematics and Theoretical Physics, University of Cambridge). The experiments were setup to create stratified turbulence generated via an exchange flow through a long tilted duct. |
| Type Of Material | Database/Collection of data |
| Year Produced | 2024 |
| Provided To Others? | Yes |
| Impact | Dataset was downloaded 52 times only three months after being live. |
| URL | https://www.repository.cam.ac.uk/handle/1810/362347 |
| Description | Dr Hao Fu |
| Organisation | University of Chicago |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | Guided experimental work and analysis of the results / writing of the paper. |
| Collaborator Contribution | Developed a new lab experiment to investigate mixing caused by shallow cumulus clouds in the atmosphere |
| Impact | A paper to appear in J. Fluid Mech. |
| Start Year | 2023 |
| Description | Dr Rocky Geyer |
| Organisation | Woods Hole Oceanographic Institution |
| Country | United States |
| Sector | Charity/Non Profit |
| PI Contribution | Fluid dynamics insights to interpret environmental data in an estuary |
| Collaborator Contribution | Shared with me valuable field data from the Connecticut River estuary, codes for analysis, as well as scientific insights into the processes at play |
| Impact | Working on a draft of an article. |
| Start Year | 2024 |
| Description | Prof Duran Matute - TU Eindhoven |
| Organisation | Eindhoven University of Technology |
| Country | Netherlands |
| Sector | Academic/University |
| PI Contribution | I co-supervised an MSc intern for 10 months. My expertise of the stratified inclined duct experiment helped the numerical modelling of the system. |
| Collaborator Contribution | Prof Matias Duran Matute was the main supervisor, contributing direct numerical simulation expertise and project management. |
| Impact | The project resulted in a successful MSc thesis defence and results are being written up for publication. |
| Start Year | 2023 |
| Description | Prof Miles Couchman |
| Organisation | York University Toronto |
| Country | Canada |
| Sector | Academic/University |
| PI Contribution | Fluid dynamics analysis of simulation outputs in the context of estuarine turbulence |
| Collaborator Contribution | Shared and co-developed codes for analysis of massive datasets arising from one of the world's fastest supercomputers in Oak Ridge National Lab (USA) |
| Impact | Working on a draft of a paper |
| Start Year | 2024 |
| Description | Prof Steve de Bruyn Kops |
| Organisation | University of Massachusetts Amherst |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | Fluid dynamics interpretation of computer simulation output of high-dynamic range stratified turbulence |
| Collaborator Contribution | Shared massive datasets and provided access to one of the world's fastest supercomputer (Frontier, at the Oak Ridge National Laboratory). |
| Impact | Working on a draft of an article. |
| Start Year | 2024 |
| Description | Prof Yalim - Arizone State University |
| Organisation | Arizona State University |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | My research data was used by the group of Prof Jason Yalim at Arizona State University to train a neural network to reduce the dimensionality of the data, as a proof of concept to automatically predict turbulence in large datasets. |
| Collaborator Contribution | Development of a convolutional neural network to predict turbulence from my datasets, using high-performance computing. The work was done by an undergraduate intern from Arizona State University who benefited from my experimental data and insights and from the supervision of supercomputing experts locally. |
| Impact | NSF Grant proposal by collaborators, where I wrote a letter of support (outcome pending) |
| Start Year | 2023 |
