Title: Characterization of the Hellisheidi-Threngsli CO2 sequestration target aquifer by tracer testing

Abstract

Mineral sequestration is among several promising methods of CO2 emission reduction. It involves
incorporation of CO2 into a solid phase via precipitation of carbonate minerals. A prerequisite to
carbonate precipitation is the availability of aqueous metal cations and a network of porous media for
fluid flow and water rock interactions. The Hellisheidi-Threngsli lava field in SW Iceland comprises
ideal conditions for studying the feasibility of permanent CO2 storage as minerals in basaltic rocks.
Prior to the injection, detailed information needs to be gathered to delineate the CO2 injection strategy
and reservoir potential to store CO2. In heterogeneous porous aquifers, simulations and predictions of
groundwater flow and solute transport require detailed knowledge of aquifer parameters and their
spatial distribution. Tracer testing offers the possibility to efficiently investigate the aquifer between
the injection and sampling wells and to characterize the relevant aquifer properties based on effective
parameter values. Tracer tests can be performed at laboratory and field-scales with depth integrated
(two-dimensional) or multilevel (three-dimensional) set-ups, and under natural or forced hydraulic
gradient conditions. Both non-reactive and reactive tracer compounds can be used. This contribution
reviews depth integrated and natural and forced gradient tracer test methods, their fields of application
at different transport scales, the SF6 and Na-Fluorescein tracers and their applications, high resolution
multi-level/multi-tracer methods, as well as approaches to evaluate tracer experiments and to quantify
tracer transport. Finally this study reports on a forced gradient dipole tracer test conducted between
wells HN-02 and HN-04 at the Hellisheidi-Threngsli site to characterize the physical properties of the
main aquifers to answer whether tortuosity and porosity will provide enough reactive surface area for
CO2-water interaction with basaltic rocks in target zone or not. Simulation and interpretation of initial
tracer test results suggest that most of the water flows through a homogenous thick layer of low
porosity, fine-medium grained basaltic lava, with high tortuosity along the flow paths, which will
provide a large reactive surface area for water rock interactions.