Translating Laboratory Compaction Test Results to Field Scale

Abstract:

In recent studies on the surface subsidence caused by hydrocarbon recovery of the Groningen gas field, the predicted subsidence is overestimated if results of compaction experiments are not corrected by an empirical ‘upscaling factor’. In order to find an explanation for this ‘upscaling factor’, an analysis is presented of different laboratory experiments conducted by NAM on samples of the Groningen field. In the mentioned studies, the result of the 1^st loading cycle is generally used for the compaction calculations, while in the 2^nd and subsequent loading cycles a lower compaction coefficient (C_m) is observed. It is also observed that stress path has a significant influence on the measured C_m. A maximum of 25 % of the discrepancy in lab and reservoir scale compaction can likely be attributed to this difference in stress path between laboratory and reservoir. The C_m values of the 2^nd cycle compaction experiments with a stress path similar to the stress path of the Groningen reservoir are very comparable to the best-fit line used for predicting reservoir compaction. These results would imply that a 2^nd loading cycle is more representative of actual reservoir compaction.

Introduction

The production of hydrocarbons leads to a decrease in pore fluid pressure in a reservoir and consequently to an increase in effective stress acting on the rock reservoir matrix. This increase in effective stress can lead to reservoir compaction, which can have serious negative consequences such as surface subsidence, induced seismicity and well shearing (e.g. Geertsma, 1973a-b; Nagel et al., 2001; Doornhof et al., 2006; van Thienen-Visser and Breunese, 2015; Fokker and van Thienen-Visser, 2016).

The subsurface of the Netherlands contains numerous hydrocarbon reservoirs, mostly gas fields. By far the largest of these reservoirs is the Groningen gas field, located in the northeastern part of the Netherlands. The Groningen field is the largest gas reservoir of Western Europe. The gas holding layers of the Groningen reservoir are Rotliegend sandstones of aeolian and fluvial deposits (de Jager and Geluk, 2007). Gas production started in 1963 and is planned to continue for the coming decennia. Surface subsidence of the Groningen area is monitored and the maximum current surface subsidence is ~30 cm in the Loppersum area (Fig. 1; NAM, 2013).