Designing effective strategies for the remediation of soil and groundwater contamination depends on high-quality subsurface information and an in-depth understanding of ongoing physical, chemical and biological processes. Conventional site investigation typically relies on the collection of only a limited set of discrete, localized (punctual) observations, which often fail to capture the full extent of the subsurface heterogeneity. This can lead to significant uncertainty in derived conceptual site models (CSMs), which propagates into human and environmental exposure and health risk assessment and the subsequent design of remediation strategies, and is furthermore usually left unquantified.

This research project focuses on the use of geophysical methods as complementary tools to conventional site investigation methods, offering time-efficient and cost-effective means to collect high-resolution and spatially comprehensive subsurface data.  We will advance 3D subsurface modelling based on geophysical inversion by adopting a probabilistic approach and improving the integration of multi-source and multi-scale datasets. We pursue to enhance model accuracy and resolution, and to explicitly quantify the uncertainty involved in each of the steps of data acquisition, processing, analysis, modelling and interpretation. We will develop and validate novel methodologies using both synthetic and real field cases, which represent subsurface conditions with varying degrees of complexity and spatial heterogeneity, resulting from natural and/or anthropogenic processes. As a higher-level outcome, we target to establish a flexible yet robust framework for the creation of digital twins of contaminated soil and groundwater systems. Such digital twins can strengthen the numerical backbone of CSMs, thereby providing improved decision-support concerning the investigation, monitoring and remediation of contaminated sites.