Periodic Reporting for period 2 - lidBathy (Nearshore bathymetric inversion from lidars during extreme events)
Période du rapport: 2022-09-01 au 2023-08-31
The coastal science community currently lacks insights into the rapid morphological evolution of sandy beaches during storms. Sediment exchanges occurring between the subaerial and subaqueous regions of the beach under the action of storm waves are poorly understood, and as a consequence, our capacity to predict the evolution of wave-dominated coastlines remains limited. This is partly explained by the difficulty to measure the seabed elevation under broken waves, and its evolution through time, with sufficiently high spatial resolution in storm conditions. The overarching goal of the lidBathy project is to study the sediment exchanges occurring at the sea/land interface during storms by developing new approaches to remotely map the nearshore topography and bathymetry, and its evolution through time. Such mapping will be performed with lidar scanners, which are remote sensors capable of safely and directly measuring both the intertidal beach topography and the spatio-temporal evolution of ocean waves propagating and breaking in the nearshore region. As opposed to the intertidal region, the seabed elevation over the submerged region will be estimated indirectly, i.e. via so-called depth-inversion procedures. Current solutions for depth-inversion are based on linear wave theories that are not adapted to the breaking region. Developing a new approach that accurately embeds waves non-linear dynamics is the core of the lidBathy project.
The project was organised in three Work Packages (WP), each addressing one of the three objectives (O) that contribute to the overarching goal of lidBathy:
O1. Develop a universal formula inferring the water depth h from incident wave properties that can be extracted from surface elevation measurements.
O2. Develop lidBathy, a new depth-inversion tool based on lidar technology and assess its accuracy with innovative field datasets collected at Narrabeen, Australia and the Field Research Facility (FRF) at Duck NC.
O3. Analyse the morphological response of sandy beaches to storms using lidar datasets of past events at the FRF and Narrabeen-Collaroy permanent stations.
The project was organised in three Work Packages (WP), each addressing one of the three objectives (O) that contribute to the overarching goal of lidBathy:
O1. Develop a universal formula inferring the water depth h from incident wave properties that can be extracted from surface elevation measurements.
O2. Develop lidBathy, a new depth-inversion tool based on lidar technology and assess its accuracy with innovative field datasets collected at Narrabeen, Australia and the Field Research Facility (FRF) at Duck NC.
O3. Analyse the morphological response of sandy beaches to storms using lidar datasets of past events at the FRF and Narrabeen-Collaroy permanent stations.
The work conducted during WP1 was the most fundamental of the Fellowship and aimed at deriving a universal relationship between incident wave properties and the mean water depth h. The “universal” character of this relationship means that it should work within most of the nearshore region, both in the shoaling and breaking wave regions, and for a wide range of beach and wave conditions. In total, three different approaches have been investigated using existing laboratory datasets of irregular and unidirectional waves. A Boussinesq (non-linear) wave theory was found to be the most accurate for predicting waves dispersive properties across both the shoaling and breaking wave regions.
Based on this Boussinesq theory, a new depth-inversion method was developed. This new method was assessed with laboratory datasets that offer a more controlled environment compared to the field. The three datasets exhibited different beach morphologies and varying wave conditions. Overall, the proposed methodology allowed to estimate the mean water depth with unprecedented accuracy, seamlessly from the shoaling region up to the shoreline (errors typically below 5-10%). The development and finalisation of the new methodology has been achieved during the Outgoing phase of the Fellowship at the Water Research Laboratory (WRL). This work has been published in Geophysical Research Letters (Martins et al., 2023), and was presented at the International Conference on Coastal Engineering held in December 2022 in Sydney, Australia.
The last phase of the Fellowship focused on collecting field data adapt the depth-inversion tool to the case of directionally-broad ocean waves. In September 2022, a comprehensive 3 week field campaign was organised in collaboration with Katherine Brodie from the US Army Corps of Engineers at the Field Research Facility (FRF) at Duck NC. During these experiments, surface elevation measurements were collected by three lidar systems: a long-range lidar from the dune station, complemented by two 3D lidars mounted on Unmanned Autonomous Vehicles. During the following months, major complications arose and consequently delayed the data analysis. This is explained by the size of the data to be processed (1Gb per minute of data) and the three dimensionality of lidar point clouds: noise, interpolations, computation of bispectral products for the Boussinesq theory etc. The analysis of the data is still ongoing, nonetheless, preliminary results show that the methodology developed in the lab is also accurate in the field, except in regions with strong surface currents. Following these first results, efforts were made to simplify the Boussinesq-based depth-inversion procedure, especially avoid computing bispectral products on lidar data. Theoretical simplifications were performed, and a simpler formula only requiring bulk parameters was derived. The assessment of this simpler formula is ongoing work. Lastly, new deployments of 3D lidar systems were conducted in May 2023 from the Hazaki Oceanographical Research Station, an experimental pier located along the Ibaraki coast of Japan. Lidar and bathymetric data collected at this site will help elucidating the limitations of the proposed depth-inversion approaches and inform on the site-specificity of physical processes affecting the accuracy of the proposed methodology. Given that the newly-proposed methodology is still at a development stage, and its robustness (e.g. site-specific or not) remains to be explored, the objective O3 proved to be overly ambitious. Nonetheless, this Fellowship paved the way to the derivation of a universal nearshore depth-inversion method. Efforts to further develop the new depth-inversion method and apply it to study sandy beaches response to storms will be pursued as part of the Experienced Researcher CNRS project.
Based on this Boussinesq theory, a new depth-inversion method was developed. This new method was assessed with laboratory datasets that offer a more controlled environment compared to the field. The three datasets exhibited different beach morphologies and varying wave conditions. Overall, the proposed methodology allowed to estimate the mean water depth with unprecedented accuracy, seamlessly from the shoaling region up to the shoreline (errors typically below 5-10%). The development and finalisation of the new methodology has been achieved during the Outgoing phase of the Fellowship at the Water Research Laboratory (WRL). This work has been published in Geophysical Research Letters (Martins et al., 2023), and was presented at the International Conference on Coastal Engineering held in December 2022 in Sydney, Australia.
The last phase of the Fellowship focused on collecting field data adapt the depth-inversion tool to the case of directionally-broad ocean waves. In September 2022, a comprehensive 3 week field campaign was organised in collaboration with Katherine Brodie from the US Army Corps of Engineers at the Field Research Facility (FRF) at Duck NC. During these experiments, surface elevation measurements were collected by three lidar systems: a long-range lidar from the dune station, complemented by two 3D lidars mounted on Unmanned Autonomous Vehicles. During the following months, major complications arose and consequently delayed the data analysis. This is explained by the size of the data to be processed (1Gb per minute of data) and the three dimensionality of lidar point clouds: noise, interpolations, computation of bispectral products for the Boussinesq theory etc. The analysis of the data is still ongoing, nonetheless, preliminary results show that the methodology developed in the lab is also accurate in the field, except in regions with strong surface currents. Following these first results, efforts were made to simplify the Boussinesq-based depth-inversion procedure, especially avoid computing bispectral products on lidar data. Theoretical simplifications were performed, and a simpler formula only requiring bulk parameters was derived. The assessment of this simpler formula is ongoing work. Lastly, new deployments of 3D lidar systems were conducted in May 2023 from the Hazaki Oceanographical Research Station, an experimental pier located along the Ibaraki coast of Japan. Lidar and bathymetric data collected at this site will help elucidating the limitations of the proposed depth-inversion approaches and inform on the site-specificity of physical processes affecting the accuracy of the proposed methodology. Given that the newly-proposed methodology is still at a development stage, and its robustness (e.g. site-specific or not) remains to be explored, the objective O3 proved to be overly ambitious. Nonetheless, this Fellowship paved the way to the derivation of a universal nearshore depth-inversion method. Efforts to further develop the new depth-inversion method and apply it to study sandy beaches response to storms will be pursued as part of the Experienced Researcher CNRS project.
A novel depth-inversion algorithm was developed based on a Boussinesq (non-linear) wave theory that predicts the dispersive properties of shoaling and breaking waves in the nearshore. One of the key novelties of this method is its universal character: it works seamlessly from the shoaling region up to the shoreline, unlike prior formulations that required specific calibrations. The newly-proposed depth-inversion method was assessed in the laboratory, yielding unprecedented levels of accuracy in the breaking wave region (within below 5-10% typically). Innovative field datasets were then collected at Duck NC and along the experimental pier of the Hazaki Oceanographical Research Station in Japan in order to adapt the new approach to “real” lidar data and assess its accuracy in the field. This is still ongoing work, but preliminary results indicate that similar accuracy as in the lab can be achieved in the field, except in regions where strong surface currents are observed. The new depth-inversion tool has a strong potential for generating unprecedented datasets of the morphological evolution of sandy beaches during storms. It is expected that its application to existing storm datasets (e.g. Duck NC and Narrabeen-Collaroy, Australia) will greatly improve our understanding of nearshore processes during storms, in a timely manner in the context of climate change and the associated uncertainties on storminess along the world’s coastlines.