Quantification of shallow land subsidence in Groningen since the start of the gas extraction in 1963 using daily groundwater dynamics

Abstract

Land subsidence is an important environmental and societal issue in the Netherlands. Lowering of the surface-level requires additional artificial drainage to maintain the current human-designated functions of the landscape such as agriculture. Moreover, during peat oxidation greenhouse gasses are emitted to the atmosphere that contribute to global warming. The Groningen gas field area in the northern Netherlands provides a good example of a region where various subsidence mechanisms cause a differential subsidence signal. Commercial exploitation of gas since 1963 from a porous sandstone layer at 2 to 3-kilometre depth has triggered deep compression. In addition, the shallow subsurface contains fine-grained deposits in which shallow subsidence mechanisms such as compression and peat oxidation occur. Although various studies have quantified deep compression initiated by the gas extraction, no previous study has investigated shallow subsidence in the Groningen gasfield area. The main objective of this thesis was to quantify shallow subsidence in the Groningen gasfield area since 1963 using a differential subsurface saturation state based on daily modelled groundwater levels. Several research strategies were combined to achieve this goal. For the selected study area in Nieuwolda, groundwater levels were modelled from 2016 till 2019 using the one-dimensional, process-based SWAP-model. After calibration, modelled groundwater levels were incorporated into a newly developed, empirical, shallow subsidence model. In this model, daily groundwater levels are used to compute differences in the subsurface saturation state through time, enabling the incorporation of seasonal or daily groundwater dynamics in the computation of peat oxidation using an exponential decay function. Additionally, shallow subsidence since 1963 is quantified by comparing historic and more recent measurements regarding lithology and surface-level. Finally, the shallow subsurface build-up was obtained by a field study in Nieuwolda and Appingedam. The subsidence model with incorporated groundwater level dynamics indicated an average shallow subsidence rate of 0.5 cm/year between 2016 and 2019 due to peat oxidation and (reversible) compression, which is in the same order of magnitude as subsidence measurements in this research area. Comparing the thickness of fine-grained deposits in the shallow subsurface between 1970 and 2021 yielded an average shallow subsidence rate of 0.2 cm/year. Differences in regional surface elevation measurements over time point out that the shallow subsidence rate varies between 0 and 0.9 cm/year, determined by a local, heterogeneous shallow subsurface. Modelling groundwater levels using SWAP was proven to be valuable in mimicking groundwater level dynamics, since the trend, the extremes and the mean of modelled groundwater levels match groundwater level measurements.