A methodology for the interpretation of aquifer tests: Application to CO2 residual trapping experiments at the Heletz site

Abstract Estimation of trapped CO2 is essential for assessing the potential of a site for geological carbon storage. In situ residual trapping can be obtained through Residual Trapping Experiments (RTE). RTE experiments consist in performing characterization tests e.g. hydraulic, thermal and tracer tests before and after creating the residually trapped zone of CO2 and estimating residual saturation from the differences between the two tests. We introduce a methodology for interpreting residual drawdowns from hydraulic tests, and specifically those performed before and after the creation of the residually trapped zone. Martinez-Landa et al. (2013) demonstrated that the reduction of hydraulic conductivity and the increase in storativity within the trapped CO2 zone can produce early time differences that are significant. However, our interpretation is hindered by the fact that accurate measurement of early time (a few minutes) response is difficult because the large inertia of the system prevents us from rapidly establishing a controlled constant flow-rate. This is particularly true for the RTE test at Heletz, where water withdrawal during the hydraulic tests had to be performed by air-lift. To resolve this difficulty, we use the proposed methodology which avoids instabilities derived from changes in flow rates. Our approach consists of four steps: (1) filtering of natural trends in heads to ensure good definition of drawdowns; (2) transformation of residual drawdowns into constant pumping test drawdowns, by using the Agarwal or other methods, while accounting for flow rate variations during the pumping phase; (3) computation of smooth log-derivatives to prepare diagnostic plots to aid in conceptual model identification; and (4) quantitative interpretation. The application of our approach to the Heletz RTE experiment gave rise to diagnostic plots consistent with theoretical expectations and a residual CO2 saturation of about 10%.