WP1.2 Subsidence disentanglement deep

The aim of work package 1.2 is to determine the key parameters of subsidence and describe ongoing subsidence. The total subsidence signal from InSAR data on a regional scale will be disentangled by the different subsidence processes. This includes shallow (< 50 m) and deep (~3 km) processes. Shallow processes include oxidation of organic material, shrinkage of clay, compaction and extraction of groundwater. Deep processes of subsidence are driven by reservoir depletion of hydrocarbons or salt. A combination of satellite data, geological mapping and groundwater models plus and reservoir characterization will be used to optimize the key parameters of subsidence by a Bayesian inversion approach. The different processes will be spatio-temporally quantified on a regional scale.

Contributions will be made both on the methodological side of the Bayesian inversion approach and the understanding of subsidence processes. Special attention will be given to the process of compaction and the difference of engineering based compaction and compaction due to lowering of the phreatic surface level. For the Bayesian inversion approach a data assimilation procedure will be employed; Ensemble Smoothing with Multiple Data Assimilation (ESMDA). Different data assimilation schemes will be tested and compared. In a later stage the ESMDA method will be compared to a Machine Learning approach.

The PhD candidate on this work package is Manon Verberne. The work has started 1st of May 2021. The supervisor of this work package is Peter Fokker (TNO).