What if we could predict surface subsidencemore accurately?
Drilling for oil and gas often causes the surface above to subside, which has to be considered when planning and managing this kind of work. There are numerous models that try to predict soil subsidence – most of them based on linear elastic models, but some geomaterials behave very differently. TMC employeneur Jitse Pruiksma recently published a paper in which he proposes a simpler and more accurate calculation model for predicting subsidence due to compaction in sandstone.
“The problem was that subsidence at numerous oil and gas fields was not linear with the gas or oil pressure depletion. At first, the subsidence measured was very small and the process was very slow, but after a while it began to increase and the subsidence went much faster. When the subsurface behaves like that, it’s very difficult to make accurate predictions.”
This delay in subsidence was measured at multiple fields, including the Ekofisk fields in Norway and gas fields in the Netherlands. The soils at these locations were mainly sandstone and limestone. Jitse focused his research on sandstone. “Sandstone is a very interesting material because it has a porous structure. It is a slow starter when put under pressure. After a small initial elastic deformation, sandstone starts to deform (creep) at a steadily increasing rate, until it reaches a constant pace. Unlike other materials, such as metal or plastic, the volume of sandstone decreases during the creep, as gas is pushed out of the porous structure. That’s why the speed of the creep eventually slows down again, simply because it can’t be compacted any further.”
In his paper, Jitse did research on an existing model for predicting subsidence. “In the eighties, researcher Hans de Waal did a lot of experiments on sandstone. He discovered that the faster you put pressure on sandstone the stiffer it reacts. That’s very different from a linear response. This phenomenon possibly explains the measured subsidence delay in gas and oil fields. Because the production of oil and gas is equal to a relatively fast loading rate on the reservoir rock. So the response is delayed due to the stiffer behaviour, as indicated in the laboratory experiments.”
The model developed from these findings was called the Rate Type Compaction Model. But before using it you had to make several additional assumptions, which were not consistent with each other and could easily lead to wrong results. Jitse: “In my research I discovered and proved that these assumptions were not actually necessary. I found that when we used only three basic assumptions, we could describe pretty much all the sandstone deformations that we were measuring in real life. This is a huge accomplishment, and as a result we can now use the model to predict surface subsidence more accurately in the future. It’s also possible that this model could be used for other granular materials, or a mix of materials that together behave in a similar way because of the spaces between them. From a physics point of view it is very interesting to observe and explain these phenomena.”