Uncertainty reduction and risk estimation for landscape evolution models

Extreme climatic events such as heavy rainfall or periods of drought can induce changes in the landscape morphology and soil composition. In particular, they can cause mudslides, soil depletion or silting up of watercourses in the absence of suitable facilities. It thus appears crucial to anticipate and prevent these phenomena. Landscape evolution numerical models can help in the management of these risks. They aim to mimic the primary physical processes at various scales in time and space and can provide predictions for the future dynamics in hydrographic basins. Some of the main difficulties in obtaining reliable predictions are the complexity of the input parameters, such as spatial variability of soil characteristics, and the uncertainty in the physical laws upon which landscape evolution models are developed. Furthermore, observations such as rainfall and fluxes available to characterize the model parameters can be sparse in space and time. However, only a handful of studies have explored the fundamental problem of uncertainty for landscape evolution models. The objective of this PhD is therefore to develop a new efficient and relevant workflow for model calibration on available data and estimation of the future basin dynamics and associated uncertainties in relation to climatic hazards. The developed approaches will be based on existing methodologies for sensitivity analysis, optimization and uncertainty propagation that will be adapted to the context of landscape evolution modeling. The PhD will focus on the Canche watershed (Hauts-de-France) where there is a notable risk of soil depletion and mud floods, and a wealth of observations from on-going projects in collaboration with Mines Paris - PSL and IMT Nord Europe. 

Keywords: Geomorphology, landscape evolution models, soil erosion, sediment transport, extreme climatic events, model calibration, sensitivity analysis, risk estimation

• Academic supervisor    FRANKE Christine, Mines Paris - PSL et ARMITAGE John, IFPEN
• Doctoral School    ED398 Geosciences, Ressources Naturelles et Envrionnement
• IFPEN supervisor    GERVAIS Véronique, Earth Sciences and Environment Technology Division, veronique.gervais@ifpen.fr 
• PhD location    IFP Energies Nouvelles, Rueil-Malmaison, France
• Duration and start date    3 years, starting in fourth quarter 2023
• Employer    IFP Energies Nouvelles, Rueil-Malmaison, France
• Academic requirements    University Master degree in Landscape evolution modelling
• Language requirements    Fluency in English, willingness to learn French
• Other requirements    Statistics, optimization, interest for numerical aspects