The formation of a sinkhole at the surface above a salt cavern is the ultimate result of the upward migration of the void space by a collapsing cavern roof. The collapsing roof is in competition with the filling of this space by the volumetric expansion of the rubble (so-called ‘bulking’1 ). Falling rubble is known to take up more space in volume compared to the in-situ rock. This will, in case where sufficient rock is present above the cavern, ultimately fill the cavern and supporting the roof, halting the migration. In cases where a sinkhole forms at the surface, the bulking was insufficient to stabilize the cavern before the roof of the void reached the surface2 . In the Netherlands it is likely that migration to the base of the unconsolidated sediments accelerates the last phase of collapse and the formation of the sinkhole.
This study has two main ultimate aims:
1) To Gain insight into cavern collapse: a. What are the (unambiguous) indicators of a collapsing cavern? b. How much time does collapse of the entire cavern take (relevant to understand possible mitigations)?
2) To gain insight into the consequences: a. What kind of effects can be expected at the surface? b. What is the areal extend of these sinkholes? We think that a sinkhole-model, specific for the Northern Dutch situation, is needed to further understand such events.
To this end we propose three phases: A. Investigate whether the current numerical tools are suited to study large and deep caverns underlying partially soft sediments. B. Develop and calibrate such a model for a number of generalized cavern shapes and depths. C. Use this model to run a series of theoretical, but realistic endmember models. The goal is to investigate the possible parameter space in the development and effects of sinkholes from salt caverns underlying partially soft sediments. These should help the regulator to gain insights in aims (1) and (2).
The project is ciommissioned to the Cavern Closure Consortium (CCC), The project started mid 2025.