Final Report Summary - LACOSTE (Lyapunov Analysis in the COaSTal Environment)
Coastal regions provide a wide range of resources and services, and thus represent an essential environment for human activities. However, they are rapidly degrading under the pressure of anthropogenic activities and climate change. A correct management of coastal ecological resources is, therefore, vital for a sustainable development.
Horizontal transport and cross-shelf exchanges play an important role in regulating the ecological and biogeochemical conditions of coastal regions. They are primarily controlled by meso- submesoscale processes (i.e. eddies, fronts and filaments on the order of 10 km or less; hereafter (sub)mesoscale), which originates from instabilities of the large-scale ocean currents. Thus, understanding the mechanisms which regulate (sub)mesoscale dynamics and quantifying its impact on ocean transport are key aspects for further improving the management of coastal ecological resources.
The Lyapunov Analysis in the COaSTal Environment (LACOSTE) project focused on the investigation of (sub)mesoscale processes in the Gulf of Lion (GoL, Northwestern Mediterranean), and the associated horizontal mixing and cross-shelf exchanges. The project was developed within the Equipe d'Océanographie Physique Littorale et Côtière (LOPC) at the Mediterranean Institute of Oceanography (MIO), in Marseille, France. The analysis combined in-situ measurements, numerical model results and satellite observations.
1. In-situ measurements
The analysis of in-situ measurements was based on the dataset from the Latex10 campaign (1-24 September, 2010), which focused on a strong frontal feature in the western GoL. Ocean fronts occur when waters of different properties (e.g. temperature, salinity or density) collide. They are difficult to study due their small scales (1-10 km) and ephemeral lifetime (on the order of few days). Nonetheless, they have a strong influence on how particles, heat, salt and nutrients are mixed and transferred in the oceans.
Previous results from Latex10, also based on finite size Lyapunov exponents (FSLE) analysis, evidenced that the front was associated with a transport barrier (i.e. Lagrangian coherent structure, LCS) along which exchanges between the GoL and the open NW Mediterranean occurred. Within LACOSTE, the analysis of the Latex10 observations was further developed to quantify the horizontal mixing and cross-shelf exchanges induced by the front.
1.1 Horizontal Mixing
Horizontal mixing was quantified by developing a new method to retrieve horizontal eddy diffusivity from in-situ observations (Nencioli et al., 2013; for all bibliography information please refer to the "Dissemination activities" document). Horizontal eddy diffusivity is a quantity which estimates the rate at which material and energy are mixed across the front. It is a key coefficient for the large-scale models used to study ocean circulation and global climate. The new method developed within LACOSTE is based on the hypothesis that the front profile resulted from a balance between large-scale stirring and small-scale mixing. Thus, in-situ estimates of the coefficient (small-scale mixing) were obtained by combining observations of surface temperature and salinity (front width) with Lagrangian drifter trajectories (large-scale stirring).
Such estimates provide a rare and important in-situ contribution for tuning and refining the representation of small-scale mixing in numerical models, and thus for improving both regional and global climate studies. Furthermore, the method can be applied to remote sensing data, opening the opportunity for global studies of horizontal eddy diffusivity at the (sub)mesoscale. For its relevance, this study was selected for:
- "Research Spotlight" on EOS-Transactions American Geophysical Union;
- "Editors' Highlights" on Journal of Geophysical Research - Oceans;
- the INSU/CNRS website highlights;
- the Aix-Marseille University Letter (April 2014, pages 30-33).
The Latex10 campaign also included a passive tracer experiment. Within LACOSTE, I have contributed to the analysis of the observations from that experiment, which provided more traditional estimates of horizontal eddy diffusivity (Kersale et al., submitted to JGR-Oceans). The results evidenced similar values for the coefficient, further supporting the results obtained with the novel method.
1.2 Cross-shelf exchanges
The Latex10 dataset provided the unique opportunity to quantify from in-situ observations the cross-shelf exchanges induced by (sub)mesoscale processes (Nencioli et al., submitted to JGR-Oceans). The front-induced fluxes were computed through a novel approach wich combines ship-based and Lagrangian observations: surface measurements of temperature and salinity were used to identify the continental-shelf waters leaving the GoL, and the open NM Mediterranean ones intruding on it;
Lagrangian drifter trajectories were used to estimate the velocity components associated with near-inertial oscillations;
these were removed from the instantaneous velocities measured by the ship ADCP, which in turn were used to compute the surface cross-shelf fluxes.
The results evidenced that in 10 days, the observed front induced cross-shelf exchanges equivalent to 10 to 20% of the whole GoL surface waters. Thus, few of these events are sufficient to completely renew the surface water conditions in the GoL.
This confirms that, although short-lived and localized, (sub)mesoscale processes are a key factor in regulating biological and ecological conditions in coastal regions. While in-situ observations can provide a detailed characterization of specific events, numerical model simulations and remote sensing observations represent essential tools to investigate their impact over larger/longer spatial/temporal scales.
2. Numerical Models
The analysis of (sub)mesoscale dynamics from numerical models was based on a ten-year high-resolution simulation of the GoL developed at the MIO within the LATEX project. During LACOSTE, I contributed to the analysis of the model results used to investigate the impact of (sub)mesoscale dynamics in the western GoL for the period of August-September 2009 (Kersale et al., 2013).
An extended analysis of transport structures at the (sub)mesoscale over the whole GoL was started during the last part of LACOSTE. The final goal is to first identify the transport structures in the GoL by applying the FSLE analysis to the model results;
then, to highlight the dominant transport patterns and quantify the associated fluxes using self organizing maps (SOM). Maps of transport structures based on FSLEs have been already obtained. The SOM analysis will be most likely developed in the next months by further extending my collaboration with the OPLC group at the MIO.
3. Satellite Observations
As also evidenced by previous results from the Latex10 campaign, traditional altimetry measurements are currently unreliable in coastal regions. Therefore, the analysis of (sub)mesoscale dynamics from remote sensing observations followed two main approaches: develop new techniques to improve existing products; contribute to the development of new platforms of observations.
3.1 Improve existing products
I contributed to a study aimed at improving the reconstruction of 2-dimensional surface velocity fields from multi-satellite along-track altimetry observations (Bouffard et al., 2014). Different interpolation techniques and datasets were used. The accuracy of the various reconstructed velocity fields was tested via different diagnostics which included Lagrangian trajectories and transport structures from FSLE analysis. The results evidenced that despite some improvements in the general transport properties, velocity fields from traditional altimetry observations remain of limited use for the analysis of coastal dynamics and transport. Thus, the development of new platform of observations will be essential to overcome such limitations.
3.2 New platform of observations
Within LACOSTE, I had the possibility to get involved in the Surface Water and ocean Topography (SWOT) mission. The mission is a collaboration between NASA and CNES for the development of a new-generation high-resolution satellite altimeter which will provide global observations of (sub)mesoscale dynamics. The launch of the SWOT satellite is currently scheduled for Fall 2020.
Since 2013, I have been member of the SWOT - Science Definition Team (SWOT-SDT). My main contribution is to lead the design and organization of the french ocean field campaigns associated with the AirSWOT missions.
A project for a field campaign in the GoL in fall 2014 (SeaGoLSWOT) was approved by the French national fleet commission, and ship time was reserved. However, due to delays in the development of the AirSWOT sensor by NASA, the campaign had to be put on hold. The campaign would have provided observations to validate the AirSWOT measurements and, most importanlty, it would have been used to test novel sampling strategies for the direct observation of physical dynamics and physical-biogeochemical interactions at the (sub)mesoscale. Depending on the status of the AirSWOT mission, SeaGoLSWOT will be most likely rescheduled within the next years.
Horizontal transport and cross-shelf exchanges play an important role in regulating the ecological and biogeochemical conditions of coastal regions. They are primarily controlled by meso- submesoscale processes (i.e. eddies, fronts and filaments on the order of 10 km or less; hereafter (sub)mesoscale), which originates from instabilities of the large-scale ocean currents. Thus, understanding the mechanisms which regulate (sub)mesoscale dynamics and quantifying its impact on ocean transport are key aspects for further improving the management of coastal ecological resources.
The Lyapunov Analysis in the COaSTal Environment (LACOSTE) project focused on the investigation of (sub)mesoscale processes in the Gulf of Lion (GoL, Northwestern Mediterranean), and the associated horizontal mixing and cross-shelf exchanges. The project was developed within the Equipe d'Océanographie Physique Littorale et Côtière (LOPC) at the Mediterranean Institute of Oceanography (MIO), in Marseille, France. The analysis combined in-situ measurements, numerical model results and satellite observations.
1. In-situ measurements
The analysis of in-situ measurements was based on the dataset from the Latex10 campaign (1-24 September, 2010), which focused on a strong frontal feature in the western GoL. Ocean fronts occur when waters of different properties (e.g. temperature, salinity or density) collide. They are difficult to study due their small scales (1-10 km) and ephemeral lifetime (on the order of few days). Nonetheless, they have a strong influence on how particles, heat, salt and nutrients are mixed and transferred in the oceans.
Previous results from Latex10, also based on finite size Lyapunov exponents (FSLE) analysis, evidenced that the front was associated with a transport barrier (i.e. Lagrangian coherent structure, LCS) along which exchanges between the GoL and the open NW Mediterranean occurred. Within LACOSTE, the analysis of the Latex10 observations was further developed to quantify the horizontal mixing and cross-shelf exchanges induced by the front.
1.1 Horizontal Mixing
Horizontal mixing was quantified by developing a new method to retrieve horizontal eddy diffusivity from in-situ observations (Nencioli et al., 2013; for all bibliography information please refer to the "Dissemination activities" document). Horizontal eddy diffusivity is a quantity which estimates the rate at which material and energy are mixed across the front. It is a key coefficient for the large-scale models used to study ocean circulation and global climate. The new method developed within LACOSTE is based on the hypothesis that the front profile resulted from a balance between large-scale stirring and small-scale mixing. Thus, in-situ estimates of the coefficient (small-scale mixing) were obtained by combining observations of surface temperature and salinity (front width) with Lagrangian drifter trajectories (large-scale stirring).
Such estimates provide a rare and important in-situ contribution for tuning and refining the representation of small-scale mixing in numerical models, and thus for improving both regional and global climate studies. Furthermore, the method can be applied to remote sensing data, opening the opportunity for global studies of horizontal eddy diffusivity at the (sub)mesoscale. For its relevance, this study was selected for:
- "Research Spotlight" on EOS-Transactions American Geophysical Union;
- "Editors' Highlights" on Journal of Geophysical Research - Oceans;
- the INSU/CNRS website highlights;
- the Aix-Marseille University Letter (April 2014, pages 30-33).
The Latex10 campaign also included a passive tracer experiment. Within LACOSTE, I have contributed to the analysis of the observations from that experiment, which provided more traditional estimates of horizontal eddy diffusivity (Kersale et al., submitted to JGR-Oceans). The results evidenced similar values for the coefficient, further supporting the results obtained with the novel method.
1.2 Cross-shelf exchanges
The Latex10 dataset provided the unique opportunity to quantify from in-situ observations the cross-shelf exchanges induced by (sub)mesoscale processes (Nencioli et al., submitted to JGR-Oceans). The front-induced fluxes were computed through a novel approach wich combines ship-based and Lagrangian observations: surface measurements of temperature and salinity were used to identify the continental-shelf waters leaving the GoL, and the open NM Mediterranean ones intruding on it;
Lagrangian drifter trajectories were used to estimate the velocity components associated with near-inertial oscillations;
these were removed from the instantaneous velocities measured by the ship ADCP, which in turn were used to compute the surface cross-shelf fluxes.
The results evidenced that in 10 days, the observed front induced cross-shelf exchanges equivalent to 10 to 20% of the whole GoL surface waters. Thus, few of these events are sufficient to completely renew the surface water conditions in the GoL.
This confirms that, although short-lived and localized, (sub)mesoscale processes are a key factor in regulating biological and ecological conditions in coastal regions. While in-situ observations can provide a detailed characterization of specific events, numerical model simulations and remote sensing observations represent essential tools to investigate their impact over larger/longer spatial/temporal scales.
2. Numerical Models
The analysis of (sub)mesoscale dynamics from numerical models was based on a ten-year high-resolution simulation of the GoL developed at the MIO within the LATEX project. During LACOSTE, I contributed to the analysis of the model results used to investigate the impact of (sub)mesoscale dynamics in the western GoL for the period of August-September 2009 (Kersale et al., 2013).
An extended analysis of transport structures at the (sub)mesoscale over the whole GoL was started during the last part of LACOSTE. The final goal is to first identify the transport structures in the GoL by applying the FSLE analysis to the model results;
then, to highlight the dominant transport patterns and quantify the associated fluxes using self organizing maps (SOM). Maps of transport structures based on FSLEs have been already obtained. The SOM analysis will be most likely developed in the next months by further extending my collaboration with the OPLC group at the MIO.
3. Satellite Observations
As also evidenced by previous results from the Latex10 campaign, traditional altimetry measurements are currently unreliable in coastal regions. Therefore, the analysis of (sub)mesoscale dynamics from remote sensing observations followed two main approaches: develop new techniques to improve existing products; contribute to the development of new platforms of observations.
3.1 Improve existing products
I contributed to a study aimed at improving the reconstruction of 2-dimensional surface velocity fields from multi-satellite along-track altimetry observations (Bouffard et al., 2014). Different interpolation techniques and datasets were used. The accuracy of the various reconstructed velocity fields was tested via different diagnostics which included Lagrangian trajectories and transport structures from FSLE analysis. The results evidenced that despite some improvements in the general transport properties, velocity fields from traditional altimetry observations remain of limited use for the analysis of coastal dynamics and transport. Thus, the development of new platform of observations will be essential to overcome such limitations.
3.2 New platform of observations
Within LACOSTE, I had the possibility to get involved in the Surface Water and ocean Topography (SWOT) mission. The mission is a collaboration between NASA and CNES for the development of a new-generation high-resolution satellite altimeter which will provide global observations of (sub)mesoscale dynamics. The launch of the SWOT satellite is currently scheduled for Fall 2020.
Since 2013, I have been member of the SWOT - Science Definition Team (SWOT-SDT). My main contribution is to lead the design and organization of the french ocean field campaigns associated with the AirSWOT missions.
A project for a field campaign in the GoL in fall 2014 (SeaGoLSWOT) was approved by the French national fleet commission, and ship time was reserved. However, due to delays in the development of the AirSWOT sensor by NASA, the campaign had to be put on hold. The campaign would have provided observations to validate the AirSWOT measurements and, most importanlty, it would have been used to test novel sampling strategies for the direct observation of physical dynamics and physical-biogeochemical interactions at the (sub)mesoscale. Depending on the status of the AirSWOT mission, SeaGoLSWOT will be most likely rescheduled within the next years.
