Numerical and analytical modelling of wellbore storage effects in low-enthalpy geothermal well tests

In geothermal wells, build-up phases of well tests are typically executed with a surface shut-in, rather than with a downhole shut-in. Therefore, wellbore physics may play a dominant, but undesirable, role in the system's pressure response as recorded by the downhole pressure gauge. Numerical simulations were carried out to investigate to what extent these effects complicate the analysis of the well test and the deduction of the reservoir characteristics. Numerical models, supported by analytical modelling of well tests of field cases, show that significant portions of the Bourdet derivatives of geothermal well tests are expressions of physical effects in the wellbore, rather than the reservoir's pressure response. These wellbore storage effects in low-enthalpy geothermal wells are more intense than commonly reported in the literature. In addition, they may last for a very long time and vary continuously during the build-up. They cannot be represented accurately with the known analytical solutions commonly available in analytical well-test software. In many cases the early, middle and late time regions of the reservoir response on the derivative plot have only developed partially, or might be completely missing. Due to their intensity and duration, the wellbore storage effects can obliterate the true reservoir response, thereby making a reliable analysis of the reservoir characteristics impossible. This will undoubtedly lead to incorrect interpretations of the skin or reservoir properties. It is therefore strongly recommended that well tests in geothermal wells are executed using a downhole shut-in device, such that the unwanted wellbore physics are fully eliminated and the downhole pressure gauges only measure the true reservoir response.