- Host institution
- Université de Bordeaux
- Doctoral school
- Sciences and environments - ED 304
DescriptionSandy shorelines are constantly evolving, threatening frequently human assets. Sea-level rise will exacerbate coastal erosion to an amount that remains highly uncertain. Sea-level-rise driven erosion is traditionally estimated using the Bruun model. Although this model has been largely criticized for overlooking other driving processes, until recently, no alternative model has been proposed. Shoreline erosion has been typically addressed deterministically, therefore overlooking the inherent uncertainty in the entire coastal change system, including other processes not related to sea-level changes. Computationally efficient sandy shoreline change models accounting for lonsghore and cross-shore sediment transport, hard structures, sources and sinks of sediment and sea-level rise have recently emerged and are under rapid development (see CoSMoS-COAST and LX-Shore). Using the LX-Shore model and dedicated statistical methods, the project aims at developing a holistic approach to hindcast and further predict coastal evolution at 3 contrasted study sites in an uncertain context. The typical scales of interest are few to tens kilometres and decades. The following research questions will be tackled: How can one optimally characterize the contribution of each process (longshore, cross-shore, sea level rise and its components), and of their interactions, to coastline evolution with a reduced-complexity process-based model? How can one characterize the uncertainties of these processes? How do vary (in time and space) the contributions of each process to the overall shoreline evolution and its uncertainties? How are the shoreline and its driving processes likely to evolve in the future? In addition to the new fundamental and practical insights into shoreline evolution in a changing climate, this PhD project will deliver a new process-based model and statistical tools to address shoreline change and its uncertainties in a wide range of wave-dominated coastal environments.
Context and Motivation
Shoreline retreat along sandy coasts increasingly threatens human assets such as buildings or transport infrastructures. This will be exacerbated by sea-level rise to an amount which remains site specific and highly uncertain. Indeed, shoreline change projections inherit the uncertainties of future mean sea level changes and of other natural and anthropogenic processes affecting shoreline variability and trends. To date, there is no research or operational model able to hindcast, and therefore to predict, shoreline evolution on decadal timescale. Therefore, coastal erosion hazard is most of the time predicted either assuming that future shoreline change rates will be the same as in the past, or using the Bruun rule which assumes sea-level-rise driven erosion as a function of shoreface slope and sea-level rise. An additional shoreline retreat is sometimes added to account for the potential rapid storm-driven erosion. A deterministic projection of the future shoreline position at a given deadline is done accordingly. However, as sea level rises with an increasing rate and in light of the high temporal variability of shoreline change with sometimes alternating periods of erosion and accretion on the timescale of decades, the assumption of using past rates to predict the future is not reliable. The natural evolution of sandy coasts is also strongly controlled by the action of waves, currents, tides, extreme water levels, sediment availability on the continental shelf and sedimentary inputs by rivers, which interact at different spatiotemporal scales. To improve our understanding and modelling capabilities of future shoreline change, it is therefore a necessary requirement to account for all these processes and their uncertainties owing to both the inherently variable climate dynamics and the future emissions of greenhouse gases. This is a milestone towards a holistic approach to downscale from global/regional climate changes to local coastal impacts.
The overall objective of the PhD project is to develop a holistic approach that skilfully hindcasts and further predicts, in a sea-level-rise context, shoreline change for a wide range of wave-dominated sandy environments, that is, for a range of wave settings, anthropogenic pressure and geological inheritance. This approach will be based on the state-of-the-art LX-Shore shoreline model (Robinet 2017). While so far this model has been essentially used on academic cases, here we will address real sites where extensive data have been collected and where chronic erosion is a major issue and has different causes. Another innovative aspect relies in the detail assessment of the uncertainties in shoreline change, in how these uncertainties and the contribution of the responsible processes vary in time and space. An overreaching goal is to both reduce these uncertainties and provide guidelines to mitigate coastal erosion at the different sites. The primary scientific questions to be tackled are:
•How can one optimally describe and model the various processes contributing to the coastline evolution in a reduced-complexity process-based model: primarily the longshore sediment transport driven by obliquely incident waves; cross-shore sediment transport as a response to incident wave energy arriving at the coast; sea-level-rise driven shoreline retreat; coastal structures affecting the natural sediment transport pathways; source and sink of sediment such as beach nourishment and sand extraction?
•How do these processes interact, and how do these interactions cascade up to the scales and explain the overall shoreline change?
•How can one characterize the uncertainties of these processes?
•How do vary (in time and space) the contributions of each process to the shoreline evolution and its uncertainties?
•How are the shoreline and its driving processes likely to evolve in the future at the 3 study sites?
Efforts will also be devoted to the upscaling of the developed method.
Methodology and Planning
The work will rely on the joint use and further development of the LX-Shore shoreline change model (Robinet, 2017) and statistical methods (Monte Carlo methods, sensitivity analysis, propagation of uncertainty). LX-Shore uses a cellular approach to describe complex shoreline plan-shapes and can handle the presence of hard structures. The application to several academic cases showed its ability to reproduce shoreline variability on the timescales from hours to several decades with low computation times (10 years simulated in few minutes to few days). This model will be used to reconstruct past evolutions and to further provide future projections of real coasts in probabilistic and possibilistic frameworks, with the primary objectives to quantify the respective contributions of each processes and their associated uncertainties. The statistical analysis methods and the information related to sea-level rise and its uncertainties will rely, among others, on the probabilistic projection work of Le Cozannet et al. (2016), as well as on possibilistic theories (Le Cozannet et al., 2017).
In addition to academic cases, there will be three application sites: a 50-km long homogeneous sandy coast (Gironde, France), a local and heterogeneous zone (Capbreton, France) with the presence of coastal structures and an offshore canyon affecting incident wave characteristics and a massive sand nourishment (“Sand Engine”, Netherlands). These sites have been intensively monitored by the supervising team and collaborators and have been selected for their complementarity in terms of hydrodynamic conditions, spatial footprint, beach morphology, anthropogenic action and “sediment balance / imbalance” states.
The planned work program is as follows:
•Task 1: Literature review & State of the Art [Year 1], focusing on nearshore processes, uncertainty quantification, LX-Shore model, and application sites.
•Task 2: Development [Year 1 & 2]
- LX-Shore model: implemente and validate the missing functionalities to address real coasts (complex bathymetries, open lateral boundary conditions)
- Method: define the probabilistic framework, define the sampling method and the number of simulations, apply the complete method on a test case
•Task 3: Application, Analysis and Interpretation [Year 2 & 3]. For each study site:
- Collect and analyse data (waves, bathymetry, shoreline, ...) and literature review on the site dynamics
- Set up and validate LX-Shore model
- Model uncertainties of local sea-level rise and different sea level contributions
- Apply the method developed in Task 2
- Analyse and interpret the results in terms of spatial and temporal variations of the respective contributions (longshore, cross-shore, steric effect, etc.) and associated uncertainties
•Task 4: Communication and valorisation of the results (at least: 2 peer-reviewed papers, 3 conferences, code disseminated as open source product) [Year 1, 2 & 3].
Joint PhD Advisor : Déborah Idier / BRGM
Skills requiredA candidate having: • Some knowledge in Earth Sciences • Skills in programming (preferably in Fortran) and modelling • Skills in mathematical theories of uncertainties • Skills in the use of at least one of the following language/software : python, R, Matlab • Good skills in communication (oral, writing) • If possible a first experience (training) in the coastal domain (hydrodynamics and/or sediment transport and/or morphodynamics)
Your profile is eligible to apply for the PhD/Doctorate program "Make Our Planet Great Again" if:
- You have a Master's degree or you will pass a Master's degree before August 31, 2018
- You have lived in France for less than 90 days since April 1, 2016
- You are exclusively a foreign national
How to APPLY?
Please send your CV and a letter of motivation to the following contact : email@example.com
KeywordsShoreline projections Sea-level rise Wave-dominated environment Uncertainties Downscaling Sensitivity analysis
- Funding type
- Contrat Doctoral
- Funding amount
- 1350 € Net / month
Application deadline 22/04/18
Level of french requiredNone
Level of English requiredB2 (upper-intermediate)
Annual tuition fee400 € / year
You must connect to be able to display the contacts.