UNveiling dynamics of Rapid Erosion through advanced Seismic Techniques

Introduction

Floods and landslides are an intensifying hazard in the context of climate change that affected over 2 billion people worldwide between 1998-2017. In mountainous catchments, these processes play a significant role in rapid erosion contributing to landscape evolution. Yet, multiple barriers limit our current understanding of rapid erosion dynamics, such as:

• The applicability of theoretical transport laws to complex real-world systems
• Poor understanding of the mechanism behind the initiation of rapid erosion processes
• Sparsely distributed observations in time and space

Project Overview

The UNREST project will cut this Gordian knot by pioneering an interdisciplinary observational framework that integrates several state-of-the-art seismic techniques, including:

1. Spatially dense arrays
2. Seismic interferometry
3. Machine learning

This framework will be combined with physical and numerical models on a previously unexplored spatiotemporal scale.

Objectives

This will be done through three main objectives:

1. Identify precursors and magnitude indicators of impending landslides
2. Link seismic signals to complex landslide dynamics
3. Develop new tools to track unstable flood flow across multiple scales

Implementation

The project will be conducted in four active Alpine catchments in France and Switzerland. The developed methods will also be tested in an entirely different geomorphic, geologic, and climatic setting of New Caledonia, a South Pacific island.

Expected Outcomes

UNREST will unveil critical thresholds in damage development on unstable slopes and previously unobserved patterns and behaviors in complex flow dynamics.

These findings, augmented by the development of near-real-time data processing workflows, will have a transformative impact on the understanding of rapid erosion dynamics, paving the way to enhanced warning systems in mountainous catchments.